07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
ARTICLE
CREATING WORKABLY COMPETITIVE
WHOLESALE MARKETS IN ENERGY:
NECESSARY CONDITIONS,
STRUCTURE, AND CONDUCT
Peter C. Carstensen*
TABLE OF CONTENTS
I.
INTRODUCTION ..................................................................... 86
II.
THE CONDITIONS OF ENERGY MARKETS: INDUSTRIAL,
TECHNOLOGICAL AND INSTITUTIONAL ISSUES ..................... 90
A. Production of Gas and Electricity................................. 90
B. Transmission................................................................. 94
C. Distribution—Retail ..................................................... 97
D. Endogenous Institutional Conditions......................... 100
F. Some Preliminary Inferences about the
Achievement of Workable Competition...................... 103
III.
STRUCTURE ........................................................................ 104
A. Gas markets ................................................................ 106
B. Structuring Electric Markets to Facilitate
Workable Competition ................................................ 110
IV.
CONDUCT ........................................................................... 119
*
Young-Bascom Professor of Law, University of Wisconsin Law School, 975
Bascom Mall, Madison, Wisconsin 53706. pccarste@wisc.edu. An initial draft of this
article was presented at the University of Houston Law Center conference: Creating
Competitive Wholesale Energy Markets, March 4, 2005. I am indebted to the attendees at
the symposium for many helpful comments including some significant disagreements with
my perspectives. In particular, John Hilke and Mark Hegedus provided me with a
number of important insights. Andrew Wang and Lief Jorgensen have provided research
assistance. Despite all this help, any errors that remain are the responsibility of the
author.
85
07CARSTENSEN FORMATTED
86
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
A. The Problem of Access and Exclusionary
Conduct—Transmission and Pipeline Monopolies..... 123
B. Bundling, Tying, Market Manipulation and
Related Practices ........................................................ 125
C. Naked Restraints of Competition—Tacit and
Express Collusion ....................................................... 128
D. Necessary Naked Restraints ...................................... 132
E. Ancillary Restraints—Reasonable and
Unreasonable .............................................................. 134
F. Jurisdiction and Penalties—Sources and Severity .... 136
G. Conduct Remedies and the Present State of the
Market and the Law ................................................... 138
V.
PREREQUISITES FOR WORKABLE WHOLESALE ENERGY
MARKETS—A RECAPITULATION ......................................... 139
I.
INTRODUCTION
The law has long sought to protect and promote competitive
markets in many industries. Recently, it has begun the attempt
to bring competitive market processes to the wholesale markets
for natural gas and electricity. Historically, these industries both
operated as regulated monopolies, and each has significant
inherent monopoly components. Hence, moving to competition
requires careful responses to market power and its potential
abuse. Overall, there is apparent progress in the market for
natural gas, but serious problems in the market for electricity.
Even in the gas market, the limits of the reform process have left
significant opportunities to engage in strategic behavior that
have produced serious harms.
This article has two goals. First, this article examines the
differences between natural gas and electricity markets and
explains why reorientation in gas has worked reasonably well,
while electricity has proven problematic. Second, the article
draws on the learning and experience from competition law and
policy to identify the policies needed to bring about a more
workably competitive wholesale market for electric energy and
improve the effectiveness of the wholesale market in natural gas.
These suggestions are not made with an expectation that they
would be implemented in electric markets. The kind of political
commitment required to accomplish such results is impossible in
the contemporary context. Rather, the contrast between these
idealized prescriptions and actual policies suggest the potential
failure to create a market structure and rules of conduct that
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
87
would serve the public interest in competitive electric markets.
Indeed, the costs and policies necessary to bring about transition
to a workably competitive market in electricity explain the reemergence of interest in returning that industry to more direct
regulation. There is justification for that pessimistic perspective,
but, on the other hand, the potential gains in efficiency and
avoided administrative costs of workably competitive wholesale
markets should cause policy makers to do a careful balancing of
the costs and benefits of seeking that objective.
To explicate the issues, this article employs in a modified
way the conditions-structure-conduct-performance paradigm.
This well known model of industrial organization posits that to
understand the performance of markets, it is necessary to start
with the conditions under which those markets operate (supply,
demand, technology, and general institutional conditions).
Conditions in turn define the kind of structural options that are
institutionally, economically, and technologically feasible. Given
a market structure, a range of individual and collective conduct
options to produce and distribute the goods or services in
question are then possible. The performance of a market can then
be evaluated. Performance is measurable along at least four
lines: static efficiency, dynamic efficiency (progressiveness),
fairness, and interpersonal equity.1 As a matter of description,
this paradigm identifies the interaction of conditions, structure,
and conduct that produce more or less desirable performance,
depending on the variable used for measurement.2 The paradigm
also facilitates normative judgments and prescriptions for public
policy informed by a sense of the initial target and expected effect
of the intervention. If, for example, structural variations do not
seem to have much consequence or are not technologically
feasible, then the appropriate focus of policy is either on basic
endogenous conditions or on regulation of market conduct.
Hence, the paradigm is used here as an organizing device that
identifies in a relevant sequence the components of economic
activity.
The paradigm has an apparently deterministic quality:
1.
See CARL KAYSEN ET AL ., ANTITRUST POLICY 11–20 (1965).
2.
The preference of the evaluator among the performance values will strongly
influence both the evaluation and any prescription for the structure and conduct of the
market being examined. In recent times, many have opted for static efficiency as the
primary goal of the economic system. The proponents of this perspective have called this
“consumer welfare” or “allocative efficiency.” As a matter of economic history, concerns for
fairness and equity have been very strong forces in defining legal duties. Others,
including the author of this article, have taken the position that a primary goal in market
regulation should be facilitating dynamic efficiency. See Peter Carstensen, Antitrust Law
and the Paradigm of Industrial Organization, 16 U.C. DAVIS L. REV. 487 (1983).
07CARSTENSEN FORMATTED
88
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
conditions set the parameters for structure; structure is the basis
for conduct; and conduct determines performance. However, the
paradigm actually has a strong dynamic character.3 Changes in
conduct can and often do affect structure as well as the
underlying conditions. The most obvious example is when the
conduct involves research and innovation.4 Similarly,
performance can result in legal intervention in the market that
changes specific conduct. The result is a change in incentives
that can alter the relative value of different structural options.
Once the system is understood in dynamic terms, legal
intervention is more likely to effect the ultimate performance of
the market, but the effects are harder to predict because of
uncertainty about the ways in which conditions, structure and
conduct will interact. For this reason, unintended consequences
are likely to result from any intervention. As Lindblom has
argued, rational public policy proceeds in an incremental fashion
responding to the consequences of prior interventions.5
Antitrust law focuses on the structure and conduct of
markets. But it is important to recognize that the basic
conditions under which markets operate are crucial to their
functioning. While many of those conditions are exogenous,
others are the result of social and legal choices that are
endogenous to the society. Whether or not an economy has a
viable law of contract greatly influences the relative efficiency of
alternative modes of integrating production and distribution. The
weaker the law of contract, the greater is the benefit from
ownership integration—assuming a workable law of property
rights.
It follows that when making a transition from one set of
legal conditions, e.g., direct regulation of price and service, to a
market system, policy makers should closely evaluate the full
range of endogenous conditions under which the proposed market
will operate and modify them as much as possible to facilitate the
new methods of doing business. One central lesson from
“deregulation” of various regulated industries is that the failure
to pay attention to the conditions of these markets has resulted
in significant problems for the evolution of desirable competitive
conduct.6 The thesis of this article is that an appreciation of the
3.
Id. at 499.
4.
New technology can change the efficient scale of production, up or down, and the
relevant inputs.
5.
DAVID BRAYBROOKE ET AL., A STRATEGY OF DECISION 83, 83–84 (1963).
6.
See, e.g., Richard Pierce, The State of the Transition to Competitive Markets in
Natural Gas and Electricity, 15 ENERGY L.J. 323 (1994); Peter Carstensen, Evaluating
"Deregulation" of Commercial Air Travel: False Dichotomization, Untenable Theories, and
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
89
dynamics of the interaction among conditions, structure, and
conduct is essential for successful reform. Many market
conditions are endogenous to the legal-economic system that
created them. As such they will be dysfunctional or
counterproductive if the market is to operate in a radically
different way. Moreover, such endogenous conditions are ones
that the legal and social system created. Consequently, they are
amenable to modification within the constraints set by the
exogenous conditions of technology, supply, and demand.
Part II of this article reviews some of the salient conditions,
under which energy markets operate and are now expected to
operate as competitive markets. These conditions are both
exogenous and endogenous. A central observation is the inherent
difference between gas and electricity markets in a variety of
exogenous dimensions. These dimensions contribute features
that either facilitate or make more difficult the transformation of
the market process. This discussion also highlights some of the
key endogenous conditions that are amendable to change.
Part III discusses the structure of these markets. Despite
assertions to the contrary, market power is omnipresent in both
industries. The decision to rely on competition ought to have led
to the restructuring of control of the assets devoted to the
production of electric energy, its transmission and distribution,
as well as a more complete separation of gas pipeline operation
from participation in the market for natural gas. Structural
reorganization is a key element in both eliminating market
power and facilitating workably competitive wholesale markets.
Structures created for one set of conditions are not well adapted
to a market oriented system.
Part IV examines the ability of conduct regulation to
implement market competition. This part suggests that even
with current structures, it is possible to draw on experience in
enforcing antitrust and other competition laws to identify rules
and policies that might reduce the risks of opportunistic behavior
and so, potentially, induce the gradual transformation of market
structure toward a more workably competitive one. But reliance
on conduct remedies alone is unlikely to yield workable
competition in electric markets.
Part V revisits the goal of competitive wholesale markets in
gas and electricity in reiterating that changes in the conditions,
structure, and conduct under which these markets operate are
necessary to develop and retain viable competitive wholesale
markets. To achieve the long run benefits of competitive
Unimplemented Premises, 46 WASH. & LEE L. REV. 109 (1989).
07CARSTENSEN FORMATTED
90
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
wholesale markets in electricity will require major changes in the
endogenous conditions, as well as the structure and conduct of
that industry. Those changes are much more dramatic and
fundamental than those required to create workably competitive
markets in natural gas. Given the political realities of policy
development, it is a close question whether it is in the public
interest to continue to pursue a vision of competition as a central
method for the marketing of electric power.
II. THE CONDITIONS OF ENERGY MARKETS: INDUSTRIAL,
TECHNOLOGICAL AND INSTITUTIONAL ISSUES
The conditions under which energy is produced, distributed,
and sold to consumers are central to the structural and conduct
options that exist for the constituting of competitive wholesale
markets. This part reviews key conditions with a particular
emphasis on the differences between gas and electricity.
A. Production of Gas and Electricity
Natural gas production is a highly dispersed activity
involving limited economies of scale or scope. Although gas needs
to be processed to make it useful, there is little differentiation in
the product once that process is complete. Gas was initially a
byproduct of oil wells and only with the development of large
scale pipelines did it emerge as a separate energy source of
7
significance. There are vast reserves throughout the world that
can be shipped if the price is right and key investments are made
in liquification plants and receiving facilities. The discovery and
production of gas is responsive to price changes. When prices
increase, new gas comes on line. When prices decline, supply
declines as marginal producers exit the market. Moreover, given
a pipeline system, gas can be easily moved in discrete quantities
from one location to another. Thus, buyers at one end of a
pipeline can transact with sellers at the other end without the
need for elaborate balancing of the entire network of pipelines.
Such a production system is not amenable to utility rate
regulation. Indeed, in its disastrous Permian Basin decision, the
Supreme Court imposed an unworkable regime of utility price
8
regulation on gas at the well head. The ultimate, and entirely
predictable, result of this changed legal condition was chaos in
7.
See FRED BOSSELMAN ET AL., ENERGY, ECONOMICS AND THE ENVIRONMENT 432,
438 (2000).
8.
Phillips Petroleum Co. v. Wisconsin, 347 U.S. 672, 685 (1954).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
91
the gas supply market.9 Ultimately, Congress sought to restore
competition in the sale of gas, but adopted a highly regulated
transition period based on the false assumption, endorsed by the
industry itself, that supplies were very limited and new gas
would be very costly to produce.10 In fact, but entirely predictably
given the nature of the production process, modest price
increases resulted in substantial new supplies entering the
market. The result was the collapse of the regulated supply
market and significant economic problems for producers and
buyers.11 Today, gas prices are rising again because of the lack of
new production despite substantial increases in demand. New
supplies will come with changes in infrastructure, i.e., pipeline
connections to Alaska and expansion of capacity to handle
liquefied natural gas.
In the case of production of electricity, the conventional
wisdom up to the 1970s was that there were economies of scale
as generation facilities got larger and larger.12 This is still true up
to a size of 250 megawatts (MW) to 500MW in conventional
generation, but technological advances in both combined cycle,
gas fired generation, micro turbines, and fuel cells have resulted
in generation capacity with much smaller minimum efficient
scales of operation.13 The fact that the scale economies were not
continuous led to recognition that generation can have a
competitive structure.14 The prospect of such competition
stimulated innovation that has resulted in the emergence of the
newer, smaller scale efficient methods of production. Instead of a
single producer operating a single huge plant, multiple plants
with different owners are now technically feasible.
This opens a path to competition in supply if the demand for
electricity is sufficiently large relative to the scale of the entering
plant. If the available market consumes 15,000MW, and the new
entry would produce 300MW, then the addition does not
significantly alter supply in the market and competition is
possible. If the relevant total demand is only 2,000MW, then the
9.
See BOSSELMAN, supra note 7, at 466.
10.
Natural Gas Policy Act of 1978, 15 U.S.C. §§ 3341–348 (1978); See WRT Energy
Corp. v. FERC, 107 F.3d 314, 321 (5th Cir. 1997).
11.
JAMES GRIFFIN ET AL., ENERGY ECONOMICS AND POLICY 301–03 (1979).
12.
Joseph Tomain, Whither Natural Monopoly? The Case of Electricity, in THE END
OF NATURAL MONOPOLY: DEREGULATION AND COMPETITION IN THE ELECTRIC POWER
INDUSTRY 114–18 (Peter Grossman, Daniel Cole, ed., 2003); Peter Grossman, The Zenith
of the Natural Monopoly System, in id. at 98–101.
13.
INT’L ENERGY AGENCY, DISTRIBUTED G ENERATION IN LIBERALIZED ELECTRICITY
MARKETS 27–28 (2002).
14.
See James Meeks, Concentration in the Electric Power Industry, 72 COLUM. L.
REV. 64 (1972).
07CARSTENSEN FORMATTED
92
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
addition of such a plant is very significant and would be likely to
result in significant disruption of the market. Thus, the size of
the demand for which a specific plant might compete is a central
issue determining the potential for competing suppliers of
electricity. Thus, the capacity of the transmission system to
deliver power is a crucial element.15
Other aspects of electricity, however, further complicate the
basic proposition that multiple generators could exist and
compete in a market with sufficiently large demand. First,
electricity, unlike gas, cannot be stored. So as demand varies
over a day and over the year, generation needs to come on line or
be taken off line. Indeed, from an economic perspective, each
time period is a separate market because both demand and
supply conditions in each period are largely independent of those
defining other time periods. Moreover, generation facilities
themselves need to vary in their productive capacity—some are
needed short term and others long term (base load). A peak load
generator will have no effect on the market for base load
electricity, but can, despite modest production, have a significant
impact on peak load prices. To the extent that new technology
allows plants to serve different segments of demand efficiently,
this serves to expand the size of the market and make
competition more feasible.16 Hence, while the product, electricity,
is a very standardized good as a result of the legal definition
necessary for its widespread distribution and consumption,17 its
production is highly differentiated.
Second, new generation involves major sunk costs to
construct and staff a plant. Day to day markets create major
risks for investors because of the potential for fluctuating prices.
15.
FERC has on sought to create larger areas for integrated transmission, but it
has also adopted “locational marginal pricing” (LMP) which imposes separate prices for
zones or nodes based on constraints on transmission. The result is that the small markets
for power are recreated and often uncompetitive. ANDREW OTT, LOCATIONAL MARGINAL
PRICING-BASED ENERGY MARKET MODEL (Aug. 25, 2003), http://www.caiso.com/docs/
2003/08/25/200308251009179503.pdf.
16.
Specifically, combined cycle generators can ramp production up or down quickly
and with little impact on marginal operating costs. Thus, such generation can compete at
both the peak and mid-range demand levels. POWER GENERATION SERVS . DEP’T, GENERAL
ELECTRIC GAS TURBINE FOR ENTRY LEVEL TRAINING (2000).
17.
The fact that different countries have different standards for electricity
including voltage demonstrates that the uniformity of electricity in a market is a result of
overt public decision. Absent such a decision, however, it would be impossible for different
systems to exchange power and it would seriously undermine the capacity of
manufacturers of electric products to manufacture standardized goods based on a uniform
standard for electricity. Indeed, some uses of electricity in computing rely very much on
the specific cycles in the system and can be harmed if there is much deviance from the
specifications. See Conrad H. McGregor, Electricity Around the World, available at
http://users.pandora.be/worldstandards/electricity.htm (last visited Feb. 28, 2006).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
93
Generally, one would expect that investors would match long
term sunk costs with long term contracts to buy the output of
such facilities. This behavior reduces risk by assuring a market
for the production. This is particularly true when, as in
electricity, the volume of sales from a new plant may be
substantial relative to the most immediate market opportunities
for the type of power being produced. Hence, new investment in
generation can be lumpy where the relevant market is modest in
size relative to the amount of power a new plant will need to sell.
Third, the lack of storage means that at all times the system
must have a reserve source of power available. This includes both
the spinning reserve that can be turned on very quickly and
additional reserves that are standing by for use on relatively
short notice if demand changes or an existing generator must
shut down.18 The system must continue to deliver power to satisfy
demand or it will collapse, and someone must contract for,
manage, and pay for the provision of these services. The
strongest version of this condition assumes that the demand for
power is very price inelastic. This is in part a consequence of
current endogenous pricing systems that use average costs
rather than marginal costs to set rates during the day. If time-ofday pricing were available, buyers would have more incentives to
vary use in response to price as well as invest in their own
production of power to reduce or eliminate use of outside power
in periods of highest price (i.e., demand). But even if demand
were much more responsive to prices, the lack of storage would
still require back up sources.
Fourth, unlike gas, electricity will not follow a path defined
by contractual commitments; it flows along the line of least
resistance. Hence, a generator adding power to the transmission
system will serve those users along the line of least resistance
even if the contractual commitment is to another user. Given the
homogenous nature of electricity, absent other transmission
issues, each wholesale buyer must simply arrange for the
delivery of an amount of power equal to its usage for
consumption to balance production. But when and where power
is added and taken off the network can result in congestion on
the transmission system creating load pockets and isolating
generators from their potential markets. It introduces another
strategic variable in the market for electric power that is much
18.
Reserves consist of the spinning and non-spinning reserve that must be ready
within ten minutes and a quick-start reserve that can come on line within thirty minutes.
See California ISO, California ISO Glossary, http://www.caiso.com/aboutus/glossary/ (last
visited Oct. 12, 2005).
07CARSTENSEN FORMATTED
94
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
more salient than in the case of gas because of the ability to store
gas at various points along the system.
Finally the electric delivery system has to be in balance at
all times. Addition or withdrawal of power anywhere in the
system creates a need to re-balance the system immediately to
avoid overloads or deficits. The need for balance, the potential for
congestion, and the need to maintain ready reserves to respond
to sudden increases in demand or declines in expected supply
mean that a competitive electric generation market presupposes
a transmission system that is highly coordinated and designed to
handle the range of inputs and withdrawals in such a market.
B. Transmission
Transmission involves the moving of bulk gas or power to
the point of distribution for consumption. This is comparable to
shipping goods to market. For most commodities, a generic
transportation service such as trucks or railroads or ships
provide this service. In the case of trucking and shipping, the
government provides or oversees the pathway19 for the vehicles
while railroads tend to own their own lines, but were statutorily
20
constrained not to carry their own products. Thus, the means of
transportation—water, road and rail—are kept separate from the
ownership of the goods passing along these systems. The early
history of antitrust litigation in railroading showed that it was
not feasible to allow railroads to deal in their own products,
usually coal, because they had incentives to discriminate and
exploit the market.21 Thus, in competitive markets, the
infrastructure necessary for the delivery of the goods is separate
from the producer and wholesale buyer using that system.22
Historically, neither gas nor electricity transportation employed
that distinction. The integration of ownership of these
transmission systems and the commodities carried on them is not
a technological necessity, but rather an endogenous policy
choice.23
19.
Pipelines and transmission wires are similar to highways, waterways, or
railroad lines. They are the paths or mediums through or on which the actual commodity
moves.
20.
See Anti-Pass Acts, ch. 3591, § 1, 34 Stat. 584, 585 (1906) (repealed by omission
1994; formerly codified at 49 U.S.C. § 1(8) (through 1978); 49 U.S.C. § 10746 (through
1994).
21.
See United States v. Reading Co., 226 U.S. 324, 359 (1912).
22.
Highways and waterways are inherently open access systems in which the
infrastructure provider, the government, is not a significant user.
23.
The fact that some forms of organization are or appear to be more efficient
forms of organization does not mean that they are not endogenous. The key question is
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
95
Like highways, waterways and rail lines, gas pipelines and
electric transmission grids have characteristics that make an
integrated system (network) the logical form of organization.
There are significant economies of scale. With gas pipelines, a
bigger diameter pipe involves less material relative to its
capacity than a smaller diameter one. Hence, there is a clear
incentive to build as large a pipeline as necessary to serve
existing and projected demand.24 Once built, capacity can be
increased to some extent by increasing the pressure under which
the gas is moved, modest remodeling to eliminate bottlenecks,
and the development of storage areas near end users that can
receive gas during off-peak periods.
Similarly, in electricity, a single transmission system using
higher voltage lines is substantially more cost effective than
building duplicate facilities. Moreover, in the case of electricity,
the needs to provide balance and reliability argue in favor of
using a single integrated transmission system. Indeed, under
present technology, it is impossible to have competing
transmission systems that are connected to the same buyer
because of the physics of electricity.25
To facilitate workable competition, again like other
transportation modes, gas or electric delivery systems should
connect as many producers with as many customers as possible.
This yields network effect gains and implies a grid like structure
to the delivery system so that goods can move between as many
26
buyers and sellers as possible. But both the gas and electric
delivery systems were initially built primarily to deliver energy
to specific sets of buyers from specific sources. Thus, neither was
built to provide a network serving multiple places for inputs and
withdrawals in the way that highway, rail, and water delivery
systems operate. The nature of gas is such that it is relatively
easy to interconnect existing pipelines where they cross or where
whether the organization is a conscious choice of policy or whether the technology
commands that result.
24.
Of course, once a pipeline is in place, further growth in demand beyond
projections might well warrant additional entry because it is also very difficult to increase
the size of a line that is already in place.
25.
It is possible to use direct current transmission lines that have a unidirectional
flow to deliver power from a source to a buyer. Some such lines exist currently in the
United States including recently developed lines linking Connecticut to Long Island.
Another similar line links New Jersey to Long Island. On the other hand, recent efforts to
develop such a line in New York floundered on the uncertainty of the profitability of such
an investment. See Rick Stouffer, Merchant Transmission Developers See Financing
Crunch, Delay of Projects, POWER MARKETS WEEK, July 29, 2002, at 1.
26.
See Daniel Spulber and Christopher Yoo, On Regulation of Networks as Complex
Systems: A Graph Theory Approach, 99 N W. U. L. REV. 1689 (2005).
07CARSTENSEN FORMATTED
96
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
they are in close proximity. Such linkages do not create an
overall need to rebalance the entire pipeline system. They simply
allow the movement of gas from one line to another much as
traffic moves from one highway to another. Moreover, major
buyers who are end users of gas could also easily connect with
the major pipelines directly without creating an imbalance in the
system.
Another important aspect of gas is that it can be stored at
various places along a system of transmission so that variations
in demand can be balanced much more easily. Indeed, each gas
pipeline can vary the amount it transports without affecting
other pipelines. Thus, gas pipelines are more like waterways,
highways, or rail lines and can be used in varying quantities
without disrupting the entire network. This again makes it more
feasible to transact in a market because there is less need for
elaborate scheduling, balancing and backup.
The key constraint on developing new gas supplies is the
infrastructure necessary to collect, process, and move the gas to
the markets in which it will be consumed. The development of
pipeline technology within the North American continent has
facilitated the movement of gas from production to consumption.
Gas can be directed to specific destinations along the pipeline
network by simply opening or closing off lines. This facilitates
contracting between producers and consumers or other buyers.
Thus, while an integrated network is essential to a competitive
market in gas, the monopoly facility, the pipeline system, is
easily separated from the competitive market transactions that
employ this system.
Electricity grids have strong reason to interconnect because
of the economies in reliability, but they face a more difficult task.
The lack of storage, the need for backup supplies, and preserving
balance at all times require strong management of the flows of
electricity. In addition, congestion on an electric line has much
more impact on the entire system than excess demand in some
part of a gas pipeline. This also means that new suppliers or
adding individual large buyers to the transmission system will
create challenges for the overall balance and congestion in the
system. Thus, electricity requires a closely integrated network. If
a line is congested and cannot carry the power needed to a
specific destination, then all other lines connected to that line
face potential disruption as the power seeks the path of least
resistance. At the same time, the fact that there is no storage
capacity means that a ready reserve must exist at all times if the
system is to be able to deliver electricity as it is demanded. The
implication is that someone must pay for some generators to be
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
97
on standby in order to maintain the system effectively.
In order to have a workable wholesale market, the
transmission system needs to be configured so that it serves a
large demand relative to the supply from any one seller, and has
the capacity to balance supply and demand for power from a
variety of sources. Technologically, such a system is feasible, but
it requires careful design and management.
The physical configuration of electric transmission facilities
in the United States does not conform to that which a
competitive wholesale market would require. Because of the
public policy decision to permit vertical integration of the
production,
transmission
and
retailing
of
electricity,
transmission lines were built primarily to move power from
specific generation sources to specific areas of use. These focused
transmission systems were linked in order, initially, to provide
more sources of backup power and so increase reliability.
Modifying these transmission grids to serve a more open,
competitive generation and sale of power from multiple sources
involves substantial and costly reconfiguration.
C. Distribution—Retail
Like transmission, the technology for delivery of gas and
electricity imposes substantial limits on competition. Basically,
both commodities need to have a continuous connection to the
consumer to be useful. This is an aspect of the network
requirements of this system. Moreover, the costs of building
duplicate delivery systems makes such an option impractical.
Hence, physical distribution is monopolistic for most customers.
However, large volume consumers can, if properly located,
connect directly to the transmission system itself and avoid
dealing with the local distribution system. This ability, if the
buyer can legally make such a direct connection, gives such a
buyer bargaining power with the local distribution system as to
its charges for the services provided. But the vast majority of
consumers lack the ability to avoid use of the local distribution
system. Moreover, each user requires its own connection to the
transmission-distribution system which makes possible
significant differentiation among customers in terms of price and
service unless controlled by regulation.27
The only other constraint on the ability of the distribution
27.
Arbitrage is possible for immediately adjacent buyers if they can link their
delivery systems. Then the buyer with the lower prices can resell to the buyer facing
higher prices. Current regulation tends to forbid such efforts and reinforce the monopoly
of the local utility.
07CARSTENSEN FORMATTED
98
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
system to differentiate among customers (this is often called
“discrimination”) comes from the potential to employ alternative
sources of energy. In the case of gas and electricity when used for
heating or other similar energy uses, there is potential for
substitution of other energy sources. Specifically, oil and coal can
provide heat and so limit the ability of any source to control
consumers. However, for illumination, electricity has no viable
substitute. Moreover, current technology makes self-generation
uneconomic for most users in most locations. However, when a
customer has the scale to make self-generation a realistic option,
it again obtains some bargaining power. This bargaining power
advances the specific economic interest of the customer with the
capacity to negotiate. It can also affect positively the market
price for electricity by making demand more price elastic.28
The technology of delivery, therefore, sets important
constraints on both the structure and conduct that is feasible in
energy markets—wholesale or retail. Contemporary technological
constraints basically require a pipeline system for delivery of
29
gas. At the same time, it is easier to separate the physical
delivery of gas (the costs and operation of the transmissiondistribution component) from the purchase of gas itself. The
presence of numerous producers at one end of the transmissiondistribution network and a substantial number of buyers at the
other end (including a number of large consumers who can buy
directly) makes it easier to create a wholesale market in gas.
Indeed, the existence of multiple buyers is itself an important
element in the development of a workable market. The number of
buyers needs to be substantial in order to get the benefit of price
competition. Due to large industrial buyers’ ability to participate
in the gas market directly, this helps to preserve larger numbers
of buyers.
Electric technology provides alternatives to obtaining power
outside the transmission-distribution network. The potential for
distributed generation is the subject of ongoing policy
30
discussion. Small generators are feasible using a variety of
energy sources including solar panels, wind generation and fuel
cell technology. The distributed energy model calls for such
28.
Introducing more demand response to energy prices is a very important aspect
of overall public policy. This article will advert to some of the key issues related to
demand, but its primary focus is on the supply side.
29.
It is imaginable that gas could be compressed as is done with propane and
shipped by rail or barge to the user, the diseconomies of collecting, compressing,
delivering, and uncompressing make such an alternative implausible with today’s
technology.
30.
BOSSELMAN, supra note 7, at 699–702.
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
99
sources to supply power to the owner of the generator and also
contribute power to the network. The network provides the
backup and reliability functions. Distributed generation might
develop to the point where it could allow a consumer to
disconnect from the electric network and still have a sufficient
and reliable supply of power. Such autarky involves risk and
costs because each individual system would then require backup
capacity to ensure reliability and the ability to deliver variable
loads which might necessitate significant excess capacity. Thus,
even if distributed generation were to increase dramatically,
most consumers would continue to be tied to the electric network.
On the other hand, if large buyers had real time prices and the
ability to switch off the network and generate their own power,
this could create more demand elasticity. Such elasticity would
reduce the incentives to increase prices to extraordinary levels.
The current pattern of integrated generation, transmission
and retailing eliminates a large volume of business from the open
market. Moreover, the refusal to allow large industrial users to
buy electricity directly further limits the number of buyers who
can participate.31 The barriers to such participation are as much
a result of the structure of legal regulation, an endogenous
condition, as they are of the complex nature of electric
transmission and distribution. The technical needs of the electric
network system may also make it difficult for resellers to exist
independent of the delivery system. The result is that these
conditions have deterred the development of a robust wholesale
market in electricity.
In addition, there may well be significant economies of scale
in the operation of larger retail networks with respect to the
service and maintenance of such systems. If this is correct and if
effective retail competition requires operation of the distribution
system, then the potential for developing a significant number of
wholesale buyers of electricity is further constrained. In New
Zealand and Pennsylvania there are ongoing efforts to create a
retail system separate from physical delivery. Institutional
factors have combined with the technological characteristics of
electricity to make this a daunting task.
31.
Compare, e.g., Praxair v. Florida P & L, 64 F.3d 609 (11th Cir. 1995) (denying a
request to have power wheeled to a large commercial user); TEC Cogeneration v. Florida
P & L, 76 F.3d 1560 (11th Cir. 1996); see also Columbia Steel Casting v. Portland General
Elec., 111 F.3d 1427 (9th Cir. 1997), 523 U.S. 1112 (1998) (recognizing a right to get
power wheeled in for a large buyer); Prairieland Energy, Inc., 88 F.E.R.C. ¶ 61,153 (1999)
(rejecting effort by University of Illinois to create a specialized “wholesaler” to provide it
with access to power).
07CARSTENSEN FORMATTED
100
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
D. Endogenous Institutional Conditions
Prior to the reforms of the 1990s, pipeline companies
engaged in two activities: they provided delivery services, and
they bought gas at the wellhead or collection point and resold it
to local gas utilities. There was no technological reason for
constituting the relationship in this way. Rather, it reflected an
endogenous, political decision to allow these companies to be
merchants rather than common carriers.32 In contrast,
responding to the anticompetitive conduct of Standard Oil,
Congress had long required that oil pipelines operate as common
carriers.33
What merits particular attention in terms of the potential
to construct a workably competitive wholesale gas market is that
the structure of ownership was a result of conscious legal choice
among a set of viable options. On the other hand, major gas
pipelines were largely unintegrated into either production of gas
or its retail distribution. Historically, these companies purchased
gas from producers and resold it to retailers. This organization
made it much easier to move from a regulated market system to
a more competitive one because the changes required in the
fundamentals of the business were fewer. The two key changes
were interconnecting existing pipelines and changing the identity
of the buyer/owner of the gas on the pipelines.
Unlike gas, the great bulk of electric generation was and still
is sold directly to retail customers of the vertically integrated
firms that combine generation, transmission and retail. Some
small distributors did buy power, but they were closely
integrated with the large generators through long term contracts.
In addition, the large integrated companies swapped power to
balance their loads and cover short term demand peaks. Prior to
deregulation, therefore, there were few active, broad,
transactional markets for electricity. As a result, electricity
required much more development of market institutions as well
as more complex technological organization in order to give
consumers access to the more competitive potential of electric
generation.
In electricity, the institutional organization at least since the
1920s was one of vertical integration from generation to delivery
32.
RICHARD PIERCE, THE EVOLUTION OF NATURAL G AS REGULATORY POLICY ,
NATURE RESOURCES AND ENVIRONMENT 53–85 (1995), reprinted in part in BOSSELMAN,
supra note 7, at 458–64.
33.
34 Stat. 584 (codified at 49 U.S.C. § 1 to 27 (1976)); see William A. Mogel & John
P. Gregg, Appropriateness of Imposing Common Carrier Status on Interstate Natural Gas
Pipelines, 25 Energy L.J. 21 (2004).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
101
to the final customer.34 Ironically, in light of contemporary policy,
the regulatory revisions of the 1930s sought to consolidate
electric utilities on a regional basis through the Public Utility
Holding Company Act (PUHCA) that in turn assumed the
validity of a regulated monopoly model for electricity production
and distribution.35 As a consequence, the wholesale markets for
electricity were fragmented and limited. Only in the late 1970s
did development of independent generation sources create a
larger and more active market in electricity.36 This led to a new
appreciation of the potential to manage supply without direct
government regulation. But the technological constraints on
electricity transmission and distribution made this a complex
task. In addition, regulators and legislators in this period did not
fully appreciate the challenges involved in converting these
markets from ones overseen by performance regulation to ones
governed by workable competition.
The legal framework within which deregulation was to occur
creates another important institutional constraint.37 Originally,
administrative responsibility to oversee the production,
transmission and distribution of energy resided in local units of
government because of the need for access to streets to lay pipes
and build electric distribution systems. Unhappiness with the
results of such regulation led to control by state agencies that
regulated entry, exit and pricing. The emergence of interstate
markets in electricity that, under Commerce Clause constraints,
the states could not regulate caused the entry of the federal
government into regulating interstate, wholesale power
transactions.38
Because Congress was filing a regulatory gap and did not
want to preempt any state regulation, it drew a distinction
between wholesale and retail power that may have been sensible
at the time.39 However, the distinction is highly artificial in the
contemporary world in which large consumers are a treated as
34.
Robert L. Bradley, The Origins and Development of Electric Power Regulation,
in THE END OF N ATURAL MONOPOLY: DEREGULATION AND COMPETITION IN THE ELECTRIC
POWER INDUSTRY 43, 45–48 (Peter Grossman & Daniel Cole, ed. 2005).
35.
Public Utilities Holding Company Act of 1935, 15 U.S.C. § 79 (2005); see
Bradley, supra note 34, at 64–70 (repealed by the Energy Act of 2005).
36.
Public Utilities Regulatory Policies Act, Pub. L. No. 95-617, 92 Stat. 3117
(1978).
37.
See Richard J. Pierce, Jr., The State of the Transition to Competitive Markets in
Natural Gas and Electricity, 15 E NERGY L.J. 323 (1994).
38.
See Pub. Util. Comm’n of Rhode Island v. Attleboro Steam & Elec. Co., 273 U.S.
83 (1927); Bradley, supra note 34.
39.
See 16 U.S.C. §§ 791(a), 824–829(h) (2005).
07CARSTENSEN FORMATTED
102
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
“retail” customers40 while the purchases by small resellers qualify
as wholesale transactions.41 Moreover, to the extent that an
integrated firm produces power for resale to its own customers
that generation is outside the authority of the federal
regulation.42 On the other hand, if a state separates generation
from distribution and retail sales, as California did, it loses
jurisdiction over the generating facilities within the state even
when they primarily or exclusively sell to distributors in that
state. In addition, most regulation that can affect final demand
rests with the states. This is a highly dysfunctional separation of
regulatory authority in electricity.
Moreover, the regulatory process itself created a separation
for utilities from ordinary market procedures and rules. A
particularly troublesome example is the use of the filed rate
doctrine. That doctrine holds that when a regulatory agency has
set a rate it can not be challenged in a collateral proceeding in
which a court determines that some other rate would have been
the appropriate one.43 The doctrine had its origins in concern for
discrimination among customers of rate regulated enterprises.44
If some buyers could get a refund based on such claims, they
would have a cost advantage over their competitors. Moreover, as
a matter of administrative process, the regulatory authority
should provide the forum for the review and evaluation of any
claims that prices were unreasonable. It alone could offer a
comprehensive remedy. These policy arguments cease to have
much relevance when competition sets the prices in the market.
The agency overseeing the market has deferred to a market
process to set prices, yet the sellers still seek to claim that the
resulting prices are “filed” rates and so not reviewable in any
collateral way.45 In market facilitating regulatory systems such
as those governing securities, commodities and livestock, the law
authorizes victims of unlawful manipulation to sue directly for
damages.46
40.
See, e.g., Prairieland Energy, Inc. 88 F.E.R.C. ¶ 61,153 (1999).
41.
See, e.g., FPC v. S. Cal. Edison Co., 376 U.S. 205 (1964).
42.
N. States Power v. FERC, 176 F.3d 1090 (8th Cir. 1999), cert. denied, 528 U.S.
1182 (2000).
43.
See, e.g., Pub. Utility Dist. No. 1 of Snohomish County v. Dynegy Power Mktg.
Inc., 384 F.3d 756 (9th Cir. 2004), cert. denied, 125 S. Ct. 2957 (2005); California v. FERC,
383 F.3d 1006 (9th Cir. 2004).
44.
Keogh v. Chicago & Nw. Ry. Co., 260 U.S. 156 (1922).
45.
See, e.g., cases cited supra note 43.
46.
See Securities Act of 1933, as amended, 15 U.S.C. § 77(a) (2005); Securities
Exchange Act of 1934, as amended, 15 U.S.C. § 78(a) (2005); Commodity Futures Trading
Commission Act of 1974, 88 Stat. 1389 (codified at 7 U.S.C. § 1); Packers and Stockyards
Act of 1921, 42 Stat. 159 (codified at 7 U.S.C. § 181 (2005)).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
103
The lack of collateral review combines with the traditional
limits on the sanctions that the regulatory agency has with
respect to approved rates that turn out to be unreasonable.47
Again, the theory is clear enough. The agency regulates the rates
at the time they are imposed, including a right of protest by
buyers, hence there is little need for strong sanctions against
unreasonable rates since they are unlikely under a prior review
system. When prices are set in an unregulated market, the risks
of manipulation and strategic conduct are very different and the
role of sanctions as deterrents to such abuse takes on much more
significance. The lack of effective and timely sanctions make it
harder to have a fair and open market process because the
participants have less incentive to behave appropriately. Thus,
the regulatory institutions governing electricity at both the state
and national levels are artifacts of their time and of the
conceptions of the tasks that such agencies were to perform.
In contrast, natural gas could not be shipped until long
distance pipeline technology was developed in the period around
World War II.48 At that point, natural gas replaced coal gas in
local systems. From the outset, gas required a more pervasive
federal regulatory scheme because the primary transactions were
interstate. As a result, the statutory jurisdictional provisions
conferred somewhat greater scope to the authority of the federal
regulator.49 This took on much greater significance at the point at
which it was necessary to transform market conduct to facilitate
greater use of competition. But even in this industry the filed
rate doctrine constrains the sanctions that can be imposed.
F. Some Preliminary Inferences about the Achievement of
Workable Competition
The foregoing discussion shows that any attempt to create
and police a workably competitive wholesale market in electricity
is fraught with challenges. The capacity to engage in strategic
conduct on the part of many participants inheres in the
fundamental conditions that define the technology, supply and
demand characteristics of electricity. Further, the institutional
framework within which that industry has evolved for more than
100 years has until recently not sought to nurture the kind of
structure or conduct that would facilitate competition. Indeed,
much of the legal institutional framework, as well as the
47.
16 U.S.C. § 824 (2005).
48.
See BOSSELMAN, supra note 7, at 438.
49.
15 U.S.C. § 717 (2005); Transcon. Gas Pipeline Corp. v. State Oil & Gas Bd. of
Miss., 474 U.S. 409 (1986).
07CARSTENSEN FORMATTED
104
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
ownership of key assets, are very maladapted to developing
competition without major changes.
The natural gas industry provides a useful and instructive
contrast. Its inherent characteristics have continually facilitated
some workably competitive production markets. Moreover, the
technological constraints are much less binding. As a result,
regulators within the framework of existing legal systems and
ownership structures can move that industry toward workable
competition in wholesale, i.e., large volume markets with greater
ease and assurance that success is possible.
The next two parts will discuss in more detail the kinds of
structure and conduct that are necessary to overcome market
power and market abuse in both industries. Both analyses will
repeatedly return to the difference in basic conditions between
gas and electricity.
III. STRUCTURE
The essential structural prerequisite to creating and
retaining workably competitive wholesale markets in energy is a
substantial body of buyers and sellers that can interact through a
50
delivery system that is as neutral as possible. Both buyers and
sellers need access to a number of potential transactions on the
other side of the market. Large numbers dilute the impact of any
particular transaction, and limit the incentive and capacity to
engage in strategic conduct either to exploit the market or
exclude parties from the market. Technology sets some limits to
the feasible numbers, but the legal conditions of the market may,
as in the case of energy, unambiguously have shaped existing
structures. So long as the marketing of electricity and gas was
regulated, those structures made less difference because conduct
and ultimate performance was subject to direct controls over
price and service. With the change to greater reliance on market
forces,
the
existing
structures
can
be
seriously
51
counterproductive.
50.
Ferry County PUD, Northwest Utilities Will Try Again On Grid Agreement…,
http://www.fcpud.com/news.htm (last visited Feb. 28, 2006). Commentators have differing
views about the minimum number of buyers and sellers necessary to create a workably
competitive market. It is, however, indisputable that increasing the numbers on both
sides increases the capacity and probability that the market will be robustly competitive.
51.
FERC Manages to Please Nobody on Rate Pancaking Issue, at
http://www.platts.com/Magazines/Platts%20T&D/News%20Archive/120704_4.xml
(last
visited Jan. 17, 2005). The RTO policy of FERC seeks to create larger market areas for
electricity by eliminating rate pancaking–each separately owned part of the transmission
system charging a separate rate. The congestion in the system and the use of LMP pricing
for transmission service make markets small again.
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
105
The basic lessons from antitrust law’s concern with market
structure are that some structures are conducive to exploitative
and exclusionary conduct. When a single firm dominates a
market, it has strong incentives both to exploit its customers or
suppliers and to exclude new entry. Markets with few
competitors are prone to tacit or explicit collusion that can raise
prices or exclude marginal competitors. Successful collusion is
much more feasible when there are only a handful of firms that
must cooperate to exploit the market.
In addition, barriers to entry are important to the longer
term survival of such exploitative arrangements. If new entry is
easy and quick, then even if there are only a handful of firms in
the market, they will have less incentive to raise prices because
of the threat of entry. Similarly, investments in exclusionary
conduct are unlikely to have much effect. Thus, the basic idea is
that markets should have as many competitors as is
technologically feasible and as few barriers to entry as conditions
will permit.
The vertical integration of markets raises further concerns.
Although some have argued that the only relevant measure of
power is the horizontal share of a market,52 a more realistic
evaluation shows that vertical linkages can greatly affect the
durability and strength of market power. Vertical organization
can create significant barriers to entry especially when one level
is a bottleneck with substantial capacity to favor or disfavor
unintegrated firms with respect to access to supply or demand.53
Thus, the structural prescriptions for creating workably
competitive markets is to deconcentrate each stage to the
greatest extent possible and vertically disintegrate the industry
if there are serious bottleneck type problems.54 If deconcentration
is not feasible for some stage for technological reasons, but that
stage presents serious bottleneck problems, then ownership of
that stage needs to be so constituted and regulated that the
52.
See, e.g., Robert Bork, Vertical Integration and the Sherman Act: The Legal
History of an Economic Misconception, 22 U. CHI. L. R EV. 157 (1954).
53.
In the context of partially regulated industries, there is the further recognition
that a firm with a monopolistic position in the regulated part of the market may find
vertical integration very attractive because such integration may well allow the firm
indirectly to exploit its monopoly power. See William Baxter, Conditions Creating
Antitrust Concerns with Vertical Integration by Regulated Industries –‘For Whom the Bell
Doctrine Tolls’, 52 Antitrust L.J. 243 (1983).
54.
Expanding the geographic or product dimension of a market is an alternative
way to dilute the power of firms. Specifically, if transmission grids in electricity can be
expanded and congestion eliminated, this adds capacity to the expanded market and
thereby dilutes the market power of all participants.
07CARSTENSEN FORMATTED
106
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
incentives to exploit the market are reduced or eliminated.55
Against this backdrop, it is possible to identify the kinds of
structures that should be employed to achieve workable
competition.
A. Gas markets
In many respects, the move to wholesale gas markets
required only modest structural reorganization. There were
already a substantial number of producers and wholesale buyers
because the pipelines were not vertically integrated into retail
distribution in any significant degree. Although the pipeline
companies often owned substantial gas production, there were
also a large number of other producers so that this vertical
integration did not seriously interfere with the transition of a
more active wholesale market. Indeed, the primary changes
required were in the identity of the wholesale buyer and the role
of the pipeline company. By separating the provision of pipeline
service from the purchase and sale of the gas sent through the
pipeline, it is possible to take a dramatic step toward a
competitive wholesale market in which the primary wholesale
buyers are either distribution companies or large consumers.
Such buyers can act directly in the market or can use the services
of agents (merchants).
Achieving the result of workably competitive markets,
however, required a response to the fact that pipeline owners
were the major buyers of gas. Given rate control over the charges
for pipeline services, if these firms remained in the merchant gas
business, they would have a structural incentive and opportunity
to manipulate their pipeline capacity available to other
merchants.
The FERC imposed two structural remedies. First, it
required separation between the two aspects of the pipeline
business.56 The theory is that these two units, despite common
ownership and joint interest in profit maximization, would
behave independently. This is a very naive idea and reminiscent
of some of the early twentieth century conceptions that underlay
the divestiture remedies in Standard Oil and International
Harvester.57 In those cases, the government required the return of
55.
Cf., Verizon Communications, Inc. v. Law Office of Curtis Trinko, 540 U.S. 398,
509 (2004) (discussing the role of the FCC in regulating access to monopoly components of
the telephone network).
56.
See United Distribution Co. v. FERC, 88 F.3d 1105 (D.C. Cir. 1996) (upholding
in general FERC Order 636).
57.
See United States v. Standard Oil, 221 U.S. 1 (1911); United States v. Int’l
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
107
stock in the affiliated companies to the original owner. But, in
the case of Standard Oil, the owners were basically the
Rockefeller group. As a result, the Standard Oil combine retained
ownership coordination into the 1930s or later.
Second, FERC in effect attempted to divest and disperse
ownership of pipeline capacity itself by requiring the pipeline
owner to sell transferable rights to use the pipeline to buyers,
sellers, and speculators.58 The basic price had to be capped to
ensure sales, but within that range the pipeline owner could lock
in its operating revenue by the sale of such access rights. The
risk that the right would lose value moved to the potential user of
the capacity. This meant that a market in capacity itself could
develop because in times of short supply buyers who owned
rights could sell them at a profit while in times of excess capacity
the rights would be likely to lose value. The significance of this
strategy is that it greatly reduced the pipeline owner’s incentive
to manipulate supply of capacity—so long as the pipeline
remained strictly a capacity supplier. Having sold all or most of
the right to use its capacity, the owner’s interest was in providing
as cost effective a transportation service as possible because its
reward was in the difference between its costs and the
contractually guaranteed revenue.
This system also reduced the need for extensive rate
regulation. Basically, a market price should exist for different
levels of pipeline service that would measure the value of those
services to customers. Given a cap on the price the pipeline could
itself charge and a requirement that it sell capacity for a price
under that cap, this combination reduced the need for more
focused rate setting and provided a useful indicator of demand
for pipeline services. Whenever the rates reached or exceeded the
cap price, that was a signal of the need for more capacity or for
other solutions to the problem of providing supply to the area
that the pipeline in question served.
Unfortunately, the FERC also allowed pipeline owners to
continue to be merchants of gas making sales for delivery along
their affiliated pipelines.59 As a result, it did not eliminate the
structural incentive to manipulate the supply of capacity. The
most notorious case involved El Paso’s purchase of most of the
capacity on its pipeline to California during the period of rising
60
gas prices and excessive electric prices. El Paso used its
Harvester Co., 214 F. 987 (D. Minn. 1914) (consent decree).
58.
See United Distribution, 88 F.3d at 1149.
59.
See Tenneco Co. v. FERC, 969 F.2d 1187 (D.C. Cir. 1992).
60.
See ATTORNEY GENERAL BILL LOCKYER, ATTORNEY GENERAL’S ENERGY WHITE
07CARSTENSEN FORMATTED
108
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
ownership of pipeline capacity to restrict the flow of gas to
California. The artificial shortage allowed El Paso to raise its
resale prices substantially above what it had paid for the this
gas. This illustrates both the incentive to exploit capacity
constraints and how integrated ownership provides stronger
incentives to engage in such activities including various capacity
constraints created by reductions in capacity of the pipeline as a
result of putative repair and service needs.
El Paso’s conduct constituted an abuse of its monopoly
pipeline position. The FERC’s only response was to impose
penalties. In addition, the victims have gotten some
compensation from El Paso,61 but El Paso retains both its
merchant and pipeline business. The structural lesson that
antitrust should teach is that the appropriate remedy would be to
terminate El Paso’s right to act as merchant of gas at least to
customers that its affiliated pipelines serve.
The situation is roughly similar to that in the motion picture
industry in the 1930s and 1940s when it was the dominant
source of entertainment. The distribution of motion pictures was
integrated with the exhibition of pictures. The result was that
the major distributors coordinated their sales, supported each
other’s first run exhibition theaters thereby excluding
competition and entrenching their position in the production and
distribution of motion pictures. The antitrust response was to
order a vertical dissolution of the industry to separate the
distribution business from the exhibition.62 With large changes in
the technology for entertainment, those commands may now
seem irrelevant, but it is likely that the growth of television and
the development of other alternative means of distributing video
material were in fact facilitated by this vertical disintegration.
Thus, the lesson for the gas industry from antitrust is that
fuller separation of ownership between traders/buyers of pipeline
capacity and the ownership of the pipeline itself would have
served the goal of ensuring a workably competitive market
structure. Such a separation is a modest intervention and would
be relatively easily implemented by requiring a spin off of the
merchant capacity of the pipeline company to sell gas for delivery
63
through the affiliated pipeline.
PAPER (Apr. 2004) at 58, available at http://ag.ca.gov/publications/energywhitepaper.pdf.
61.
Id. at 62.
62.
United States v. Paramount Pictures, 334 U.S. 131 (1948).
63.
A total ban on pipeline participation in merchant activities may be appropriate
because of the risk, similar to that in the motion picture industry to reciprocal acquisition
of capacity control. If pipeline operators have no stake in the wholesale market for gas,
their economic incentive should be to seek maximization of the usage of the pipeline.
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
109
A closely related structural issue is the vertical integration
of pipeline ownership and electricity production.64 Here the
problem, as in the motion picture case, is that the pipeline can
constrain access to supplies with which to compete with its
generator. Implementing such a strategy would be contingent on
identifying a way to exploit the resulting market power. For
example, if all generation is compensated at the marginal price,
then a baseload generator will make additional profits if it can
drive up the prices of the marginal generators. Control over the
price or quantity of gas can be an effective means to accomplish
this goal. Alternatively, if the pipeline owner is allowed to own
the capacity on such a line, it might withhold supply and cut
back its own baseload generation to create an artificial price for
gas to fill the needs of intermediate and peak load generators.
To retain workably competitive gas markets strict merger
enforcement is also essential. The focus of concern should be as
much on the buying side of the market as the selling side. As
discussed earlier, workably competitive markets require
substantial numbers of buyers and sellers. As the market moves
toward a more concentrated structure on either side, the danger
of distortion grows. Indeed, buyer power can be a real problem
even with relatively modest market shares.65 Several recent
antitrust cases have made this point as has some empirical work
on the effects of mergers generally.66 If the buyer’s market share
increases and such a buyer has the capacity to store gas so that it
can vary its purchases, it would have the capacity to influence
market prices downward.67 The advantage to the buyer in doing
this would arise only if the buyer can use cheap gas in reserve
priced at the prevailing market rate. In such a situation, the
buyer would be able to raise the price of the gas resold and make
a higher profit. Although the cross linkage of pipelines reduces
the capacity to engage in such strategic conduct because sellers
are not confined to the buyers along the line of the pipeline
64.
See, e.g., United States v. Enova Corp., 107 F. Supp. 2d 10 (D.C. Cir. 2000).
65.
See Albert A. Foer, Introduction to Symposium on Buyer Power and Antitrust,
72 ANTITRUST L.J. 505 (2005).
66.
Toys “R” Us v. FTC, 221 F.3d 928 (7th Cir. 2000); Knevelbaard Dairies v. Kraft
Foods, 232 F.3d 979 (9th Cir. 2000); Todd v. Exxon Corp., 275 F.3d 191 (2d Cir. 2000);
Edward C. Fee & Shawn Thomas, Sources of Gains in Horizontal Mergers: Evidence From
Customer, Supplier, and Rival Firms, 74 J. FIN. E CON. 423 (2004). In energy markets,
particular buyer power concerns exist with respect to Entergy, large vertically integrated
utility serving Arkansas, Louisiana and parts of Texas and Mississippi.
67.
Cf., Knevelbaard Dairies, 232 F.2d at 979 (Kraft allegedly sold cheese on the
exchange in order to drive down the prices it paid for cheese in off exchange transactions);
see also, WILLARD F. MUELLER, ET AL., WIS. D EP’T OF AGRIC., TRADE & CONSUMER PROT.,
CHEESE PRICING : A STUDY OF THE NATIONAL C HEESE EXCHANGE (1996).
07CARSTENSEN FORMATTED
110
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
system to which they connect directly, still, as buyers become
larger factors in the market, they have greater capacity and
incentive to engage in such strategic conduct.
The further implication of the potential for adverse
competitive consequences from the combined impact of vertical
relationships between pipeline ownership and merchant sales
and between pipeline ownership and electric generation is that
other kinds of mergers need strict scrutiny. In particular,
combinations among pipeline operators that would also retain
substantial merchant capacity increase the risk of manipulation
of pipeline capacity (a regulated element) to advance the
unregulated merchant interest. In the case of railroads, the
abuse of such positions has led to a total ban on a railroad
carrying its own goods. Absent such a ban, the merger review
process should look critically at how the structural change
through merger would affect the incentive and capacity to
manipulate markets. The long run goal of public policy ought to
be complete separation of pipeline ownership from the use of the
pipeline either as a supplier to an affiliate of the owner, i.e., a gas
fired generation plant, or as a merchant selling gas to end users.
B. Structuring Electric Markets to Facilitate Workable
Competition
The technological characteristics of electricity create greater
challenges for the development of workably competitive market
structures. On the one hand, generation, given the scope of the
transmission system, has the potential to have a competitive
structure. However, entry and exit into generation is not easy.
This has important implications. Entry into the production of
electricity involves a very substantial investment that can be
amortized only over a significant time period. This presents
significant risk to the entrant and its financial backers. The risk
is exacerbated if the entrant has no guarantees of a firm market
for at least a significant part of its production. Thus, a process
that involves longer term contractual commitments from
appropriate buyers is the way to avoid delays and encourage
independent entry into the market.
Absent such a system for managing entry, the most likely
entrant would be an established retailer of power that expands
generation to satisfy its own projected needs. Such vertical
expansion, however, undermines the goal of creating a workable
wholesale market. It is for this reason that the much-maligned
California plan had a requirement that the existing retailers of
power had to sell off at least fifty percent of their generation
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
111
capacity.68 While other aspects of the plan nullified the gain from
developing a wholesale market inherent in such a requirement, it
does illustrate the kind of structural policy a competitive goal
requires. On the other hand, one of the biggest mistakes in the
California plan was the failure to authorize either the system
operator or retailers to enter into long term contracts for
baseload power. The absence of such contracting authority
dramatically increased the risk and disincentive to build new
capacity in the market.
Second, transmission is not currently configured to provide a
good match to the growth in competitive supply. Moreover,
unlike a highway, waterway or pipeline, an electric transmission
system has to be tightly integrated with both supply and
demand. Electricity is not storable, and demand fluctuates
substantially. Additionally, the transmission system must be in
balance and have reliable backup to respond to sudden changes
in either supply or demand. This means that a transmission
system employed as a means of delivering power from
competitive suppliers to wholesale buyers must have a great deal
of residual capacity to avoid congestion.69 In addition, some
central authority must exist to ensure reliability and balance
within the system as a whole. The more different sources of input
and removal from the system, the more complex that process
becomes.
Further complicating the transition of the transmission
system to one that could accommodate a more competitive supply
context is the physical layout of the system, as well as its
ownership. Existing transmission lines primarily serve the
integrated owners’ interest in moving power from its generators
to its customers. This system needs redesign and added facilities
to make it amenable to a more open system of supply and use.
The fact that vertically integrated firms own vital segments
of the transmission system means that without some change in
68.
New York and Maine have also sought vertical disintegration without the
problems that arose in California. ME. REV. S TAT. ANN. tit. 35-A, § 3204 (West 2005)
(Maine electric industry restructuring statute); N.Y. PUB. SERV. COMM. OP. NO. 96-12 at
65 (1996), available at http://www.dps.state.ny.us/ (New York administrative agency
opinion stating policy preference for divestiture under broad public utility consumerprotection statutes); see The Great Sell Off, ENERGY ECONOMIST, March 1, 1998
(discussing vertical disintegration in New York and California); see also Del Jones, States
Take Varied Routes to Energy Deregulation, USA TODAY, Feb. 1, 2001, at 3B (reporting
study by Center for the Advancement of Energy Markets praising electricity deregulation
in New York and Maine).
69.
Increased real time pricing and the development of cost effective self generation
capacity would create more price elasticity on the demand side and so could reduce the
need for overly generous transmission capacity.
07CARSTENSEN FORMATTED
112
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
ownership or control there is a conflict of interest between the
owner’s self-interest as an integrated producer and seller of
electricity and what a neutral operator of a transmission system
would do. So long as the control over balance, reliability, and
basic operation remained in the integrated owner, the potential
for manipulation of access is significant. While conduct-oriented
measures might ameliorate the tension, it should be obvious that
changing ownership of the transmission system is the most likely
way to eliminate the incentives for strategic conduct and provide
a centralized approach to the planning and operation of an open
access transmission grid.
Separating ownership of generation from other activities and
creating workably competitive generation markets requires
special attention to the problem of short-term market power that
arises in times of congested transmission or overall high demand.
In such contexts, generators with only modest overall shares can
acquire a great deal of market power and an incentive to game
the system by withholding electricity to raise the average price
for the rest of their sales. This is rational conduct regardless of
the gain to other generators although cooperation among
generators in a market may facilitate this kind of exploitation.70
Only by restricting the kind of generation capacity a firm may
own that serves any separable market area can the incentive to
withhold power be constrained.
Such a limit on generation capacity in a market should not
be confused with a limit on overall ownership of different types of
generation. PUHCA, now repealed, sought to consolidate utilities
into a single geographic area based on notions of efficiency within
a vertically integrated, regulated market context. A workable
competitive market, on the other hand, should avoid regional
concentrations. If there are economies of scale in managing a
larger number of generation facilities involving a range of
technology, then the law should not foreclose the opportunity to
own multiple plants. But such combinations should not occur
within the same geographic market.
This structural analysis points to vertical disintegration of
ownership of generation, transmission and the wholesale
purchase of electricity, either for resale or use by large buyers.
Moreover, this idealized structure calls for horizontal
disintegration of the wholesale buying side of the market if
70.
The degree of coordination among generators is an open question in California.
See Jacqueline Weaver, Can Energy Markets Be Trusted? The Effect of the Rise and Fall of
Enron in Energy Markets, 4 HOUS. BUS. & TAX L.J. 1, 68 (2004), available at
http://www.hbtlj.org/content/v04/v04weaverar.pdf (last visited Sept. 26, 2005).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
113
necessary to establish a number of buyers so that no one is
powerful enough to exercise monopoly power. Similarly, the
generation side needs to be so structured as to limit the potential
for market power on that side of the market.
What also emerges from this structural view is that the
separately owned and operated transmission system must take
on (and charge for) the obligations of maintaining reliability and
balance as well as reconfiguring the system in order to make it
more open to entry by new suppliers. Indeed, the transmission
system owner could also be the owner of local distribution if that
would create a more seamless system of moving power from
generation to the ultimate users, and if retail sales could be
separated from the provision of system services.
While the first step is to make transmission system operator
a separate business from either generation or downstream
buyers (large users or retailers), that would simply move the
market power to this bottleneck. The key structural question is
whether it is possible to design an ownership for this entity that
would significantly constrain its incentives to exploit its
position.71 The ownership of the system should ensure equal
access for all users, i.e., generators and buyers, and should have
incentives to eliminate congestion by investing in expansion and
reconfiguration of transmission capacity.
One strategy is to have all stakeholders, i.e., generators and
72
retailers/buyers share, in ownership of the entity. The model is
that of a quasi-cooperative. Such an enterprise has different
motivations because it is essentially established to serve its
members’ interests. If no member or group of members gets a
strategic advantage from controlling key decisions, then all
members will have an incentive to seek to make the jointly
owned facility operate in as efficient and mutually helpful way as
possible. Thus, if the potential gain from strategic conduct is
sufficiently limited, each stakeholder is better off having the
enterprise operate efficiently. Indeed, such an enterprise has an
incentive to overbuild its facilities and operate in other ways at
the high end of reliability because that is in the mutual interest
71.
The conduct analysis will examine the use of transmission rights, which seem to
be primarily a conduct oriented remedy. It is possible to make such rights more durable
and so impose on electric transmission a regime similar to that used in gas pipelines.
Unless the owner of the transmission facility is strongly disfavored in such an allocation
of rights, however, such a policy is unlikely to create the appropriate incentives to
eliminate congestion.
72.
See TRANSMISSION ACCESS POLICY STUDY GROUP, EFFECTIVE S OLUTIONS FOR
GETTING NEEDED TRANSMISSION BUILT AT A REASONABLE COST (2004), available at
http://www.tapsgroup.org/sitebuildercontent/sitebuilderfiles/effectivesolutions.pdf
(last
visited on Sept. 30, 2005).
07CARSTENSEN FORMATTED
114
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
of all participants.73
This is an application of the essential facility concept from
antitrust law. The law of essential facilities teaches that when a
monopolist controls a resource that is necessary for upstream or
downstream activity by its competitors, then, assuming access is
feasible, it may not unreasonably exclude its competitors from
access. The classic illustration was the St. Louis Terminal
Railroad that controlled all the rail routes across the Mississippi
River and through St. Louis. These routes were essential to
connect east and west railroads. A group of rail lines owned the
Terminal Railroad and imposed very high charges on their
competitors for the use of the system. This gave the owners a
significant cost advantage over their rivals. The Supreme Court
ordered that all railroads serving St. Louis be allowed to
participate in ownership of the Terminal Railroad.74 Hence,
exploitation by virtue of high prices or restricted access would be
a wash because each line would pay in and then get back the
overcharge. In addition, but not mentioned in the opinion, the
revised ownership of the terminal railroad should have induced
its management to engage in providing service on the best terms
possible to all members. Once strategic advantage was
eliminated, the goal of such a linkage in the rail system is to
provide good service and have the capacity to provide all service
needed by its stakeholders.
Wisconsin has already applied the essential facility concept
75
to the transmission lines within the state. It required all electric
utilities that owned such lines to transfer them to a newly
created corporation, American Transmission Company (ATC).
These utilities got back stock in ATC. In addition, major buyers
of power including cooperatives, municipal providers, and
distribution companies that did not have major transmission
assets were authorized to buy into ATC thereby providing it with
working capital. The board consists of executives from both
buyers and sellers of power. ATC is actively expanding and
revising the transmission system within the state to make it
73.
The Averch-Johnson effect in regulated monopolies might actually work to the
benefit of creating a workable wholesale market if the owner of the transmission system
saw gain only from regulated rates based on capital investment. Harvey Averch & Leland
L. Johnson, Behavior of the Firm Under Regulatory Constraint, 52 A M. E CON. REV. 1052
(1962). In such a context, the owner would have an incentive to “over” invest in capacity
in order to earn greater revenue. Moreover, from the perspective of stakeholders, the
relatively minor increase in cost would be helpful in avoiding interference with the
market for power.
74.
United States v. Terminal R.R. Ass’n of St. Louis, 224 U.S. 383 (1912).
75.
See WIS. STAT . § 196.485 (2005).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
115
more serviceable to all users.76
Finally, one needs to look at the buying side of the market.
The basic problem is that distribution of power is currently
inextricably linked to its sale to consumers. Moreover, because of
the institutional history of integrated power systems, the retail
power sellers tend to serve very large blocks of customers. The
implication of this fact is that there are relatively few buyers of
power for resale. This structural fact increases the risks and
uncertainties for sellers of power in any regional market. There
are, as discussed in the next Part, some conduct oriented
responses to this problem, but the question to be considered here
is whether structural responses might not provide better and
more durable solutions.
The two main structural considerations include the absolute
size of the retail operations and the integration of ownership and
control of the distribution networks themselves with the retailing
of power. To create a workable buying market, one could
restructure the retail side of the market by breaking up the
retailers into smaller units that retain sufficient size to be able to
achieve any necessary efficiencies. A probable objection is that
management of the physical assets of a distribution system may
entail more significant economies of scale; moreover, the
distribution system needs to be closely integrated with the
transmission system and increasing the number of distribution
systems may well create additional coordination problems.77
The foregoing considerations suggest that in order to have a
workably competitive market structure on the buying side and
retain the benefits of an integrated distribution system,
separation of ownership and control of the physical assets of the
distribution system from the business of retailing may be in
order. The distribution system would, like a rail line, pipeline, or
highway, provide the physical means for delivery of the product,
but the sale of the product itself would be in the hands of a third
party—the retailer. The advantage of this system is that retailers
76.
See American Transmission Company, http://www.atcllc.com (last visited Feb.
28, 2006); see also, Douglas Houston, User-Ownership of Electric Transmission Grids:
Toward Resolving the Access Issue, REGULATION, Winter 1992, at 48–57 (advocating user
ownership of regional transportation systems).
77.
See J. Stephen Henderson, Cost Estimation for Vertically Integrated Firms: The
Case of Electricity, in ANALYZING THE IMPACT OF R EGULATORY CHANGE IN PUBLIC
UTILITIES 75, 90–91 (Michael A. Crew ed., 1985) (concluding that electrical distribution
networks have “substantial economies of scale” and are natural monopolies). But see
Adonis Yatchew, Scale Economies in Electricity Distribution: A Semiparametric Analysis,
J. APPLIED ECONOMETRICS, March/April 2000, at 187, 202 (concluding that economies of
scale exist but do not reduce cost-per-customer when the distribution network grows
beyond 20,000 customers).
07CARSTENSEN FORMATTED
116
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
would be able to compete for customers in the downstream
market and could operate in several geographic areas thus
increasing their volume without being assigned a dominant
position in any one retail market.
An additional way to expand the wholesale market is to
authorize major consumers of electricity to buy directly from
generators and obtain delivery through the transmission system.
Such sales were a subject of competition between the local
distribution companies and cooperatives and the integrated
generators.78 As with natural gas, increasing the number of
buyers who deal directly with generators increases the depth and
strength of the wholesale market for electricity. Of course, the
inherent needs of the transmission system may make it more
difficult for such buyers to deal directly with suppliers.
The concept presented here involves first isolating the
primary source of inherent market power—the transmission and
distribution systems—and identifying ways to neutralize the
incentive to exploit that power. Second, the concept involves
restructuring both generation and wholesale buying of power to
create a market context that would support and nurture a
workably competitive market. The electric industry can in these
ways be converted to a structure that is much more conducive to
market processes.79
When we compare the idealized structure of a market based
electric system and the observed system, it is obvious that the
gaps are formidable. At the center is the transmission system
question. FERC’s preference is for the creation of an Independent
System Operator (ISO) or a Regional Transmission Organization
(RTO) to act as an independent operator of the transmission
system without any underlying change in ownership of or
responsibility for the day-to-day management of transmission
80
facilities. This approach is fraught with difficulties because the
operator lacks the rights that go with actual ownership,
including the right to reconstruct and develop the physical
78.
See, e.g., Town of Concord, Mass. v. Boston Edison Co., 915 F.2d 17 (1st Cir.
1990); Am. Elec. Power Co. v. City of Mishawaka, 616 F.2d 976 (7th Cir. 1980).
79.
One can find a rough analogy in the can case. United States v. Am. Can Co., 87
F. Supp. 18 (N.D. Cal. 1949). In that case, the dominant can makers also owned key
patents for can closing machines. They would only lease their machines and insisted that
lessees (canners) had to buy most or all of their cans from the patent holder-can maker.
The antitrust court ordered both an end to this practice and required that the machines
be sold at modest prices. As a result, the canners were then able to buy from all can
makers. The basic idea is that the monopoly power has to be dissipated and separated
from the activity that permits exploitation of the power; see also James McKie, The
Decline of Monopoly in the Metal Container Industry, 45 A M. ECON. REV. 499 (1955).
80.
See Notice of Proposed Rulemaking, 68 Fed. Reg. 24,679 (May 8, 2003).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
117
system. FERC has not insisted on restructuring of ownership of
generation or transmission, but has allowed horizontal mergers
and the sale of generating facilities back to retailers thus
increasing vertical integration, and has not sought to create a
large body of buyers of electricity to stimulate the demand side of
the market.81 In short, it has failed to recognize that a market
based system requires substantial structural changes at all three
levels of the traditional market of electric power.
This failure to deal with structure parallels the failure of
Congress to impose a systematic reorganization of
telecommunications markets in 1996.82 At that time, Congress
decided that there could be competition in local service. This was
not contemplated by the 1981 settlement of the monopoly case
against AT&T.83 Congress wanted to open access to local
landlines, but failed to separate ownership of the physical assets
from the operation of local service. The result has been a series of
problems as entrants into landline service were subject to
84
exclusionary practices by the incumbent providers. The saving
grace of telecommunications in the long term is the emergence of
alternative methods of providing basic and advanced
communications. There are wireless as well as cable television
based internet systems, and, potentially, electric power lines can
be used. These alternatives limit the market power of landline
systems and make the need for restructuring less pressing.
In the case of electricity, there is much less prospect of a
technological breakthrough that would create competing delivery
systems. There are those who believe that competing
transmission systems are technically and economically possible.
Such an innovation would require a very large capital investment
and would seem to raise the prospect of duplicate and
overlapping systems. The experience with competition in another
capital intensive industry, cable television, suggests that, in fact,
such competition is not practical. Absent such a major
technological transformation, competitive markets in electric
power will be best served by restructuring of the industry so that
it fits with the market oriented model that a competitive
wholesale market in power requires.
As in the case of gas markets, it is also important to control
81.
As discussed earlier, part of the problem is that FERC lacks clear authority to
impose some of these conditions. See Joseph Kelliher, The Need for Mandatory Electric
Reliability Standards and Greater Transmission Investment, 39 U. RICH. L. R EV. 717
(2005).
82.
Telecommunications Act of 1996, Pub. L. No. 104-104, 110 Stat. 56 (1996).
83.
See U.S. v. AT&T, 552 F. Supp. 131 (D.D.C. 1982).
84.
See Trinko, supra note 55.
07CARSTENSEN FORMATTED
118
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
electric company mergers.85 The overall lesson from merger
enforcement is that it is much easier to retain workably
competitive markets than to restore them. In the case of
electricity, one of the greatest problems is the acquisition of
generation facilities by retail distributors. This clearly
undermines the goal of creating workable wholesale markets.
This is particularly important in the context of a current
structure that includes a great deal of existing vertical
integration. Second, mergers that increase regional concentration
in either generation or distribution create serious problems for
the goal of competition. Once again the effect of such mergers is
to create localized market power at least in peak times or periods
of congestion. Such market power in turn allows these firms to
coerce either suppliers or customers to give favorable treatment
or exclude new entry. Because the goal of market regulation is to
create workable wholesale competition, mergers that undermine
that goal should be denied even if conventional market analysis
would not condemn them.86
Finally, as in the gas case, mergers that combine either gas
and electric distribution or allow vertical integration between gas
and electric generation raise serious competitive concerns. The
linkage of gas and electricity means that only one major local
retailer of energy exists in the market area. This makes the
development of viable alternative wholesale customers for any
energy product very difficult. Similarly, the linkage between gas
supplies and electric generation create the incentive for strategic
conduct that can distort prices and exploit customers.
The lesson of more than 100 years of antitrust efforts to
police markets is that the less concentrated the market structure,
the more competitive markets are likely to be. Applying that
lesson to particular situations requires that technological
conditions be respected. However, it should be clear that allowing
increased concentration or vertical integration in electric
85.
Such mergers have generally proven unsuccessful in economic terms. See John
Becker-Blease et al, Mergers and Acquisitions as a Response to the Deregulation of the
Electric Power Industry: Value Creation or Value Destruction? (Social Science Research
Network,
Nov.
2004),
available
at
http://papers.ssrn.com/sol3/papers.cfm?
abstract_id=625083.
86.
A significant legal policy problem with contemporary merger enforcement policy
at both agencies such as FERC and the antitrust law enforcers is the limited set of likely
consequences deemed sufficient to justify a challenge. As the discussion in this part
shows, harmful competitive consequences can arise in ways not covered by conventional
policy. The statutory standards of effect on competition and the public interest would
easily accommodate these concerns. The failure to develop better merger policy is a result
of administrative and judicial failures. Cf. Peter Carstensen & Nina Questal, The Use of
Section 5 of the Federal Trade Commission Act to Attack Large Conglomerate Mergers, 63
CORNELL L. R EV. 841 (1978).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
119
markets is going to undermine the ability to create and maintain
workable wholesale electricity markets.
IV. CONDUCT
A central premise of much of the current effort to create
wholesale markets in energy is that regulating the conduct of
existing participants in these markets, regardless of the
structure of those markets, will suffice to bring about workable
competition. This view may rest on one or both of two
questionable assumptions about industry participants. First,
because changes in conduct feed back to structure directly and
indirectly through changes in the basic conditions of the market,
it is possible that mandated changes in conduct will cause the
conditions and structure of wholesale markets to be transformed.
Second, there may be a belief that if market participants are told
that they must compete that they will do so regardless of their
economic self-interest and the opportunities for strategic
conduct.87 As Part III argues, the second assumption is naive
given the incentives created by existing structures and
opportunities for strategic conduct. The first assumes either very
strong feedback effects that could only occur if very draconian
conduct rules existed or that the structure and conditions of
energy markets do not require much in the way of modification.
Not only are current and most possible conduct rules
insufficiently draconian, but it would also be seriously wrong to
assume that only minor adjustments in structure are needed
with respect to electricity, although such a view is not entirely
implausible with respect to natural gas.
Workably competitive markets require rules and regulations
that facilitate their operation. It is an often overlooked, but
fundamental, fact of economic activity that markets need to be
constituted and facilitated by adequately enforced legal and
regulatory commands if they are to fulfill their roles. Viable
energy wholesale markets need to be reliable, not unduly subject
to manipulation, competitive, and accessible to buyers and
sellers. When shaping regulations for a workable market,
regulators must consider the market’s structure and the strategic
conduct feasible and rational for market participants. But when
market conditions and structures are significantly inconsistent
with workable competition, conduct oriented remedies are
unlikely to bring about such a result. Hence, if public policy relies
87.
It is at least arguable that FERC holds this belief based on its market behavior
rules and proposed rules. See Notice of Proposed Rulemaking, 68 Fed. Reg. 24,679 (May 8,
2003).
07CARSTENSEN FORMATTED
120
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
primarily on conduct rules to induce workably competitive
wholesale markets, it is unlikely to succeed with respect to
electricity, but it has a better chance in natural gas.
Transparency in price and quantity of sales is very helpful in
assuring fair and efficient dealing in markets with competitive
structures. Transparency allows buyers and sellers, including
firms on the margin, to enter or exit the market or expand or
contract production. The key characteristic of such markets is
that neither buyers nor sellers have individual capacity to affect
the market price. All are price takers.
In markets with structures involving high concentration on
either the buying or selling side, transparency can become a tool
for either collective action among a few dominant firms or for
unilateral manipulation of the market if a buyer or seller has
market power in some context.88 For such markets, as FERC
proceedings on market power mitigation and standard market
design demonstrate, complex conduct regulation is the only
possible way to get any of the benefits of competition. But as an
increasing number of commentators have observed, the results
seem to be a costly and fragile system that has yet to produce any
of the benefits of competition.89
Exacerbating the lack of structure or conduct in electricity
oriented toward an active transactional market are the
underlying conditions of electric transmission operation. Because
flows respond to physical laws and not the Uniform Commercial
Code, and because the entire system needs to be closely
controlled to ensure balance and reliability, both structure and
conduct evolved to address those needs within the framework of a
regulated and long-term contractual market. Only after these
structures were in place has the law sought to create a workably
competitive market. In such a context, there are large transition
costs. Moreover, change has to go forward on a trial and error
basis. Market participants have an incentive to game any system
to maximize their profits. As those opportunities emerge, it is
essential to fix them by revising the rules governing conduct. But
the underlying conditions and structure, exogenous and
endogenous, are likely to frustrate this effort at least in
electricity.
A particularly striking characteristic of the efforts to create
88.
See, e.g., United States v. Container Corp. of Am., 393 U.S. 333 (1969); Proposed
Modification of Existing Judgment in United States v. Gen. Elec., 42 Fed. Reg. 17,004
(1977).
89.
See AM. PUB. POWER ASS’N, R ESTRUCTURING AT THE CROSSROADS : FERC
ELECTRIC POLICY R ECONSIDERED (2004); Hullihen Moore, Competition: The Wrong Goal,
39 U. RICH. L. R EV. 739 (2005).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
121
markets for electricity has been the heavy reliance on short-term
transactional markets. While the special characteristics of
electricity require more last minute purchases because of
fluctuation in demand and the inability to store it, the failure to
rely on longer term contracts for baseload and even for
reasonably predictable peaks in demand is curious. Economists,
regulators and politicians seem to think in terms of a
transactional market when they talk about competition and so
converted abstract models into legislative and administrative
rules.
The bias toward transactional markets created additional
problems for entry and growth of producers of power in many
regions.90 The fact that they could not get long-term contracts
meant increased risk of lost sunk costs. At the same time, the
vertically integrated firm had a clear advantage because it could
self-contract for supplies. Moreover, without vertical separation
so that all distributors of power had to buy most or all of their
supplies, it was probably not feasible to create the kind of context
that would have induced all participants to find ways to make
the market work. It is notable that there is a strong movement in
all wholesale markets, but especially in electricity, to return to
using longer term supply contracts.
Workably competitive markets require rules of conduct that
are effectively enforced. Without rules, any market will
degenerate into a lawless war of economic attrition. Indeed, rules
for the conduct of markets are the key first step in developing
markets whether based on transactions or contracts. The rules
can come from private market regulation, as was the case in
securities and commodities markets in the 1800s and early
1900s, or from direct regulation as illustrated by the Securities
Exchange Commission and the Commodities Futures Trading
Commission that currently oversee the operation of those
markets. What is essential is that there be a relatively neutral
rule maker with the capacity to modify rules in response to
experience and the ability as well as the will to enforce the rules.
As Tom McCraw has argued, the SEC is a remarkable
illustration of such an agency and something of an anomaly in
the experience of American economic regulation.91 The rules
governing complex market conduct need to be market specific
90.
It is possible that state retail choice regimes create pressure for short-term
markets, because the entities serving retail customers in such regimes could see their
customer bases shifting frequently, which might make longer term contracting more
difficult.
91.
TOM K. MCC RAW, THE PROPHETS OF REGULATION (1984).
07CARSTENSEN FORMATTED
122
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
and go beyond the limited bounds of antitrust law. The goal of
such rules is to facilitate an open, competitive, and efficient
market for the good or service being bought and sold. Such
regulation should be compatible with antitrust law,92 but by its
nature it is a necessary first step after which antitrust can
provide further enforcement against specific types of conduct.
The policies and experience of antitrust also provide useful
guidance to those developing such rules of conduct.
Antitrust law is itself a limited factor in policing such
markets.93 Antitrust law essentially assumes that the competitive
market is the norm in any context subject to antitrust law. Only
firm specific conduct that violates that norm given the ordinary
practices of the market fall within the purview of antitrust
challenges. For example, extracting a monopoly price from a
market is not in itself unlawful under antitrust law.94 Such
conduct may prove that the firm has a monopoly and so will
permit judicial review of unilateral conduct by such a firm that
has significant exclusionary effect or even involves unlawful
methods of exploitation. Finally, antitrust law is enforced
through the courts that lack the ability and capacity to write
detailed administrative rules (or modify them) to facilitate
competitive markets.
Despite the limited role that antitrust law is likely to play in
directly regulating the conduct of participants in the wholesale
market for energy, antitrust law can provide important insights
into the competitive issues that need to be addressed and
provides some guidance on the kinds of sanctions that may be
effective in deterring violation of the rules.
This discussion cannot develop the specifics of the market
facilitating regulation that must exist to organize wholesale
markets in gas and electricity. That task would be enormously
daunting as pending and past rule making proceedings at FERC
95
demonstrate. Hence, the focus of this discussion is on the
insights that antitrust law and competition policy can contribute
to the ongoing process of developing and enforcing market
facilitating regulations. It addresses six aspects of wholesale
market constituting regulations and their enforcement to which
the policy insights of antitrust law can make contributions. The
92.
See Gulf States Utils. v. Fed. Power Comm’n, 411 U.S. 747 (1973).
93.
See Darren Bush & Carrie Mayne, In (Reluctant) Defense of Enron: Why Bad
Regulation Is to Blame for California’s Power Woes (Or Why Antitrust Law Fails to Protect
Against Market Power When the Market Rules Encourage Its Use), 83 O R. L. R EV. 207
(2004).
94.
See Trinko, supra note 55, at 407.
95.
See, e.g., Fed. Energy Regulatory Comm’n, 18 C.F.R. § 35 (2002).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
123
bottom line, however, is that it is extremely unlikely that conduct
oriented regulation can, on its own, remedy the problems of
market power and its abuse in electricity. The prospects for gas,
in contrast, are better.
A. The Problem of Access and Exclusionary Conduct—
Transmission and Pipeline Monopolies
Monopoly law is primarily concerned with exclusionary
conduct that entrenches and protects a monopolist from
competition. Antitrust’s fundamental vision of monopoly is that
one will not prove durable so long as it is subject to the
“centripetal and centrifugal” forces of the economy.96 This is a
dynamic understanding in which market power emerges, is
challenged, and dissipates again. Hence, the law condemns
unreasonably exclusionary conduct. Such behavior includes
bundling sales of goods so as to disadvantage a competitor with a
more limited line,97 entering into contracts that require either
suppliers or customers to refuse to deal with actual or potential
competitors,98 and refusing to make goods available in ways that
allow competitors to create attractive and competitive packages.99
The problems of exclusion are well recognized in energy
markets and subject to a variety of conduct oriented
interventions. A central insight of competition law is that the
best remedy is to create market contexts in which there is little
or no incentive to engage in such conduct. It is for this reason
that the previous section emphasized structural responses that
would reorganize the industries so that most participants would
have limited market power and the transmission/distribution
controllers would have no incentive to use the inherent power in
those elements of the market.
The central regulatory problem in electricity, given the
current structure of integrated ownership of transmission,
generation and retailing, is that the firms controlling all levels
have no incentive to upgrade transmission facilities so long as
they can use the bulk of the available capacity. While more
effective, after-the-fact sanctions against unjustified exclusion
might reduce the incentive, given the complex technological
issues involved in operating a transmission system, many
96.
Standard Oil v. United States, 221 U.S. 1, 62 (1911).
97.
LePage’s, Inc. v. 3M, 324 F.3d 141 (3rd Cir. 2003) (en banc), cert. denied 124 S.
Ct. 2932 (2004).
98.
Toys “R” Us v. Fed. Trade Comm’n, 221 F.3d 928 (7th Cir. 2000).
99.
United States v. Microsoft, 253 F.3d 34 (D.C. Cir. 2001) (en banc) (per curiam);
cert. denied 534 U.S. 952 (2001).
07CARSTENSEN FORMATTED
124
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
opportunities would exist to create or exacerbate congestion
problems. So long as structural integration exists it is hard to
imagine any conduct regulation that can successfully overcome
these incentives.
FERC has tried to deal with the problem by creating various
kinds of rights to firm transmission access for unintegrated
firms. To date these interventions seem not to have produced the
kind of revamping and expansion of the transmission system that
is desired. The conduct limits have not had sufficient feedback
effect on structure. Moreover, the required use of ISOs and RTOs
have resulted in dramatic increases in the costs of these services
without producing much benefit in terms of greater competition
and access to the supplies.100
One draconian remedy would be to exclude all integrated
owners from firmly committed use of their transmission
101
facilities. All firm capacity would be assigned to unintegrated
firms, and, if this number did not prove sufficient, investors
would be invited to buy such rights. To then get assured access to
its own transmission system if it suffered from congestion, the
owner would have to buy back such rights or expand the
transmission system such that all users could use it without any
congestion. The goal would be to make it very costly to continue
to operate a constrained transmission system. Where it would be
costly or difficult to expand such a system, the integrated owner
should find it economically attractive to sell the system to a
“transco” as in Wisconsin or otherwise dilute its ownership to
remove the incentives to engage in strategic conduct. Only such
an intrusive conduct remedy is likely to resolve the problem of
access on congested transmission systems.
Even if an integrated firm divested itself of its transmission,
it would still have incentives to self deal with its own generation
services so long as there was no retail competition. It could pass
on its own costs and ensure that it would incur no loss on its
sunk investments. Meanwhile, more efficient merchant
producers would find themselves at the margin of the market,
subject to strategic conduct based on the buyer power of the
integrated firm. It is impossible to ensure equal treatment of
suppliers in the context of significant vertical integration
between production of electricity and its retail distribution.
100.
See Margot Lutzenhiser, An Expensive Experiment? RTO Dollars and Sense,
PUB. UTIL . FORTNIGHTLY, Dec. 2004, at 38, available at http://www.pulp.tc/
RTO_ISO_Costs121004.pdf.
101.
Because the retail customers of these integrated utilities help build these
transmission facilities, state regulators are not likely to look favorably on such a proposal.
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
125
In gas markets, the problem is similar but the remedy is
easier. The specific problem is the incentive to use control over
capacity in the pipeline to favor the affiliated merchant arm of
the pipeline owner. Here sanctions can be directed at the owner.
In securities and commodities markets, regulators have the
authority to bar individuals and enterprises from participation in
the market when they engage in seriously wrongful
manipulation.102 Since the problems in gas are more likely to be
relatively obvious, such a sanction might provide sufficient
deterrence to eliminate the problem. Again, a structural
approach that restricted the permitted scope of activities of
pipeline owners with respect to natural gas sales would require
less costly policing.
B. Bundling, Tying, Market Manipulation and Related Practices
Monopoly law is concerned with the ways a monopolist
exploits its power to capture monopoly profits. In essence, the
law seeks to confine such exploitation to the market in which the
power exists and insists that insofar as practical the exploitation
should be done by directly pricing the power. In dynamic terms
this makes the value of the monopoly more transparent and in a
Schumpeterian world of “creative destruction” allows potential
competitors to identify the most economically attractive
opportunities for new and better products or technologies. Thus,
one of the policies supporting strict rules against tying is that
such practices both distort competition in related markets and
obscure the value of the monopoly product itself.103
The implications of the antitrust experience with tying and
bundling support the efforts of FERC to unbundle the sale of
specific components of the production and transportation of both
gas and electricity. The exogenous conditions of the gas market
create fewer problems for the easy implementation of such a
conduct requirement.
In electricity, the combination of components needed for
effective service suggest that the definition of appropriate
products for sale is more difficult and involves an expressly legal
determination of the categories or types of things to be sold and
104
who should be allowed to sell or required to buy. The goal of
102.
See Becker-Blease, supra note 85.
103.
See, e.g., Int’l Salt v. United States, 332 U.S. 392 (1947); N. Pac. Ry. v. United
States, 356 U.S. 1 (1958).
104.
This was an important concern for the D.C. Court of Appeals in the Microsoft
case because of its concern with product innovation that might involve combining two or
more previously discrete programs. A similar concern might exist with respect to classes
07CARSTENSEN FORMATTED
126
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
this kind of conduct regulation is to create a market of active
buyers and sellers. Because such markets have not generally
evolved given the past history of the electricity business, there is
a need for explicit definition, much as there are grades of grain or
livestock defined by government regulation. But the lesson from
those markets is also that definitions need to be modernized to
reflect market realities.105
Once again the presence of vertical integration of supplies
creates problems that may be insuperable. The merchant power
plant is going to be the marginal supplier for such enterprises.
When they dominate a region and have the benefit of congestion,
it is unlikely that any integrated firm will voluntarily move
toward a workably competitive supply market. The experience
with PURPA and co-generation provides case studies in
inefficient and counter productive efforts to force the purchase of
electric power by integrated utilities.106 In some respects this
“cure” is worse than the disease of market exploitation because it
basically shifts the opportunity to exploit to another set of actors.
Another variable that can strongly affect the incentives to
collude is the method of pricing the goods or services. Classic
economic theory calls for all prices to be set at the level where
supply and demand intersect. This means that the inframarginal sellers may earn economic rents (the price will exceed
their costs), but in a world of homogenous goods and no long term
contracts each seller would in fact price at that price; moreover,
no buyer would pay more than the market price. In a world of
many transactions each representing only a small part of the
total demand, this model is coherent and workable. Collusion is
difficult exactly because of the large numbers involved. On the
other hand, when the “market price” is not the result of massive
trading, it becomes vulnerable to manipulation. In commodities
markets, where most transactions occur outside the public
market, the public price can remain the basis for pricing all other
sales. In such a context, large buyers or sellers have a great
incentive to take the opposite position (buyers sell to drive down
the public price; sellers buy to drive it up). There are well
documented historic examples of this conduct in various
of electric supply. See Microsoft, 253 F.3d at 89.
105.
Agricultural Concentration and Competition Hearing, Before S. Comm. on
Agriculture, Nutrition, and Forestry, 106th Cong. (2000) (testimony of Stephen Koontz),
available
at
http://agriculture.senate.gov/Hearings/Hearings_2000/Aptil_27__2000/
00427koo.htm (regulation has failed to promote efficient and desirable transactional
markets in livestock by failing to adapt grading to needs of contemporary meat packers).
106.
See Indep. Energy Producers Ass’n v. Cal. Pub. Util. Comm’n, 36 F.3d 848 (9th
Cir. 1994).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
127
commodities.107 These examples often involved groups of buyers
or sellers acting together, expressly or tacitly, to manipulate
price. Such conduct is, therefore, uniformly condemned even if
specific regulations may not effectively control it.
In pricing energy, the central question is the interaction of
the marginal price for energy with the overall volume of sales. If
a uniform price, as in the failed California system, based on the
price of the marginal unit of power required exists in a
concentrated market, this creates significant incentives for
exploitation. Control over the price of the marginal unit raises
the net revenue of all the other units producing power for sale at
that time. Hence, the pricing model employed needs to responsive
to the market context in which sales will occur if the incentive to
manipulate is to be limited. Thus, once again the remedy is found
in finding a way to make the incentives to engage in such conduct
less attractive rather than trying to punish the conduct after it
has occurred.
The problems in gas markets that emerged primarily in the
context of California result directly from the structural decision
to allow pipeline owners to be merchants of gas combined with
the constrained capacity of the available pipelines serving the
state. While this might have been unavoidable as a transition
stage, it creates exactly the incentives to seek to control capacity
on the pipeline when manipulation of gas prices is feasible. The
lessons from competition law teach that this creates a serious
risk of market manipulation. Remedies could include controls
over such merchant activity, including forbidding transactions
involving delivery over the affiliated pipeline. But even then, the
experience of the motion picture industry, discussed earlier,
suggests a potential for collusive arrangements in which each
pipeline would favor the merchants affiliated with other
pipelines to permit collective exploitation of the latent market
power that exists as a result of the fact that pipelines have finite
capacity and the entry barriers are enormous. Another
alternative drawn from the regulation of commodities and
securities markets would be to bar violators from continued
participation in the market.108
107.
See Peter Carstensen, The Content of the Hollow Core of Antitrust: The Chicago
Board of Trade Case and the Meaning of the "Rule of Reason" in Restraint of Trade
Analysis, 15 RESEARCH IN LAW & ECON . 1 (1992); see also Knevelbaard Dairies v. Kraft,
232 F.3d 979 (9th Cir. 2000).
108.
See Robert McDiarmid, Address at the National Rural Electric Corp. Association
Meeting, Taking Stock: The Success and Limitations of Open Access (Jan. 11, 2005).
07CARSTENSEN FORMATTED
128
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
C. Naked Restraints of Competition—Tacit and Express
Collusion
Restraint of trade law also addresses both exclusionary and
exploitive conduct. The classic concern is with cartels among
buyers or sellers that fix price and limit the quantity of goods
sold. In order to survive, a cartel must also police its members
and exclude potential competition. Thus, inherently, such
conduct involves both exploitive and exclusionary elements. From
an antitrust perspective, cartels lack any justification and so are
absolutely illegal as a general matter. Exceptions, discussed in
the next section in more detail, exist when either state or federal
government has expressly or impliedly authorized a cartel to
serve some public interest. Claims of such authorization are
subject to differing degrees of scrutiny depending on the context
and era in which the case arose.109 However, the fact that such
lawful naked restraints can and will exist further complicates the
already difficult problem of policing collusive conduct.
To establish an antitrust violation two issues must be
resolved: (1) whether there is in fact a “contract, combination . . .
or conspiracy” among the enterprises and (2) whether any such
110
understanding actually “restrains” trade.
The first issue focuses on whether the conduct is the product
of unilateral decisions or interdependent conduct in which no
firm would engage without some understanding that its
competitors would do the same. Antitrust law has long
recognized that tacit collusion is as much a problem as is overt
agreement to restrain competition. The High Fructose decision
written by Judge Posner provides a good illustration of the kinds
of evidence that support a finding of collusion even in the absence
of direct evidence.111 Basically, the court considered the incentives
of the parties, the nature of the product, and specific evidence of
conduct that was inconsistent with rational unilateral decision
making. Such tacit agreements are easier to create and maintain
when few firms are involved and there is sufficient transparency
in conduct that any deviation from the understanding will result
in an immediate retaliation. Electricity is very vulnerable to such
tacit agreements because of the many time of day and generation
109.
See TODD ZYWICKI ET AL., R EPORT OF THE STATE ACTION TASK FORCE (Sept.
2003); Peter Carstensen & Bette Roth, The Per Se Legality of Some Naked Restraints: A
[Re]Conceptualization of the Antitrust Analysis of Cartelistic Organizations, 65 ANTITRUST
BULL. 349 (2000).
110.
The Sherman Anti-Trust Act § 1, 15 U.S.C. § 1 (2005).
111.
In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651 (7th Cir. 2002),
cert. denied, 537 U.S. 1188 (2003).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
129
specific markets that exist within congested areas.112 This means
that barriers to new entry exist and the generators within the
zone protected can more easily coordinate their activities.
Moreover, the fact that electricity cannot be stored means that
there is substantially more opportunity to engage in short run
market price collusion with respect to peak loads.
The second issue focuses on how the conduct impacts the
economic discretion of the actors: Does the agreement entail a
restraint? The classic example and one of great import in energy
markets is the exchange of information. Such exchanges can
operate to make a market perform in a more nearly perfect
manner as a result of more informed buyers and sellers.113 On the
other hand, some exchanges make economic and business sense
only because they imply an underlying understanding that
competition will be constrained.114
The rules organizing a structured market can either
facilitate genuine competition in the market or create significant
opportunities for strategic conduct. Both the conditions under
which the market will operate and the structure of market
participants are vital considerations. Where there are many
buyers and sellers of a relatively homogeneous product, collusion
is difficult because of the significant costs of coordination and the
incentives for individual parties to betray any collusive
understanding. There have been exceptions, however, in
numerous types of markets. Where there is a tight community of
interest in a group, and it has some means to deter members
from cheating on the agreement, collusion is possible even in
large numbers. However, such collusive understandings are
usually more open and obvious exactly because of the need for
consensus among a large group. Hence, encouraging accurate and
prompt public reporting of prices in such contexts is usually
consistent with maintaining a workably competitive market.
The analysis is different when there are significant barriers
to entry/exit, there is high level of concentration on one side of
the market, and the capacity of firms at the margin to vary their
output is limited. In such circumstances better price reporting
provides a means for tacit collusion. The number of players is
limited and so reaching understandings is easier; detection of
112.
Thus, the heterogeneous character of electric production technology and time of
day markets create a large number of contexts in which relatively few competitors can
control price. In such a circumstance, the temptation to coordinate conduct is very great.
113.
Maple Flooring Mfg. Ass’n v. United States, 268 U.S. 563 (1925).
114.
Sugar Inst. v. United States, 297 U.S. 553 (1936); Am. Column & Lumber Co. v.
United States, 257 U.S. 377 (1921); United States v. Container Corp., 393 U.S. 333 (1968);
United States. v. U.S. Gypsum Co., 438 U.S. 422 (1978).
07CARSTENSEN FORMATTED
130
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
deviation is inexpensive because of the public disclosure of the
relevant price information. Moreover, firms with larger shares
are more vulnerable to retaliation if they do deviate since price
competition will drive down their own average income. Hence,
the structure of such markets creates disincentives to compete on
price.
Antitrust law has responded to both kinds of market
situations. It has affirmatively facilitated price discovery systems
in markets with many participants. Such markets are improved
with better price disclosure. It has also successfully challenged
information disclosure in oligopolistic markets because of its
demonstrated capacity to facilitate market coordination.115 The
remedies in such cases, counter-intuitively with respect to
conventional economic models of competitive markets, have
involved requirements that obscure the prices of the competitors
in so far as that is possible. The concept is that in such markets,
uncertainty about how rivals will price creates a stronger
incentive for each firm to make its own decisions based on its
own costs. The goal is to maximize rivalry given the structure of
the market.
A serious tension exists between the administrative needs of
energy markets and the kinds of regulations that might push
suppliers to be more price competitive. To create competition,
regulations should frustrate communication and limit
transparency. But the exogenous needs of both electric
transmission and gas pipeline service conflict with that goal.
Hence, to reduce the risk of collusion, given the structure of the
markets, it would be necessary to increase the inefficiency of the
delivery systems by concealing important information from major
market participants. Indeed, it is probably not technically
possible to conceal enough information to frustrate tacit collusion
in electric markets. The potential to use conduct remedies to
overcome the structure and conditions of the market is very
limited if it exists at all. Antitrust law also allows for after-thefact liability to deter collusion. This liability includes the threat
of substantial fines and prison time.116 In addition, treble
damages can result in very substantial liability. The goal of these
sanctions is to deter collusive conduct. The general sense is that
despite the well-publicized violations, sanctions deter many firms
115.
Proposed Modification of Existing Judgment, United States v. Gen. Elec., 42
Fed. Reg. 17,004 (1977).
116.
Recent amendments to the Sherman Act increased the maximum to ten years
and some executives have received three years. United States v. Andreas, 216 F.3d 645
(7th Cir. 2000). Archer Daniels Midland itself paid a $100 million fine.
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
131
from engaging in collusive conduct. Unfortunately, the filed rate
doctrine currently precludes victims of price fixing in energy from
making damage claims. Moreover, prison is reserved for
defendants who engage in overt price fixing. In the case of
electric markets, as discussed above, tacit collusion is the most
likely kind of naked restraint on competition that will occur.
Hence, the potential for antitrust law to provide any real backup
to market regulation is very limited.
Creating a credible wholesale market in electricity faces
many challenges. The basic conditions of power production and
transmission require a more closely integrated system of
operation with a high level of communication among actors to
ensure the stability and utility of the system. The fluctuation of
demand with the consequent need for dramatic variation in
supply, including the creation of short-term market power,
creates significant challenges to any market system. Controlling
collusion is more difficult because of the inherent need for a high
level of forward communication about both supply and demand
as well as the need to make last minute purchases of power to
cover unexpected peaks in demand.
In addition, the existing structure of most markets creates
an additional set of problems for workable wholesale competition.
The continued ownership integration of generation, transmission,
distribution, and retailing creates incentives to manipulate the
wholesale price whenever it is possible for the integrated firm to
pass the higher price on to its customers. Moreover, the
necessary level of transparency and communication for safety,
balance, and reliability, assures participants that any deviation
from an understanding will be readily apparent and easily
countered.
Faced with this kind of market context and the divided
regulatory authority, the task of fashioning workable remedies
for collusion is daunting. The first and most obvious implication
from antitrust experience is that simplistic models based on the
assumptions of perfect competition have no place in this system.
Rather, the appropriate regulatory response is to find ways to
induce competitive conduct despite strong incentives to collude.
To the extent that individual retailers have an obligation to
provide both baseload and peak power to match their needs, it
may be possible that such buyers will employ a variety of
strategies to avoid being exploited by sellers. Such strategies
would consist of longer term contracts including long term peak
load commitments that would have relatively set prices. The
practice of giving the owners of such rights the ability to sell the
rights if they did not need them would further complicate the
07CARSTENSEN FORMATTED
132
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
ability of colluding producers to manipulate price in some
circumstances. Whether this can be done given the level of
vertical integration between generation and distribution outside
FERC’s authority is very questionable.
In contrast, gas markets are somewhat less vulnerable to
collusion than electric markets. First, there are more sellers and
buyers. Second, FERC rules caused some disintegration of
pipeline capacity that limits the incentive of pipeline operators to
restrict access. Third, because of the direct and indirect links
among pipelines and the capacity of traders to swap gas on
different systems as well as redirect gas among systems, a larger,
more nearly national market has emerged. Fourth, the capacity
to store gas at various places along the system again weakens the
opportunity for strategic withholding of supplies. The
combination of the inherent characteristics of the market, the
structure of the ultimate buying and selling of gas, together with
specific regulations that limit some incentives to collude, make
wholesale gas markets more nearly workably competitive.
D. Necessary Naked Restraints
Further complicating the effort to induce competition in
energy market is the need for mutually agreed to standards and
rules for the operation of the transmission and distribution
systems. Although such regulations are in fact market
constituting, functionally they are indistinguishable from
anticompetitive cartelistic restraints agreed to by firms that are
potential or actual rivals.117 The role of such agreements, as
discussed in Part II, is very substantial. Many of the firms
developing these agreements are also in a position to exercise
joint control of some aspect of the market. Moreover, while it is
possible to “just say no” to basic collusion to manipulate prices,
these interactions and resulting agreements often have a strong
basis in the inherent needs of energy markets to operate and
operate efficiently. Agreement among stakeholders is essential to
resolve many of the technical specifications for electricity and key
issues of reliability and balance. Similarly, in the gas business, a
number of operating conditions and protocols need to be agreed
upon in order to ensure that merchants, sellers, and ultimate
recipients can all operate within the system. The need for such
agreements provides a fertile ground for the parties to engage in
additional tacit collusion and to adopt unduly exclusionary or
117.
See, e.g., Fashion Originators’ Guild of Am. v. Fed. Trade Comm’n, 312 U.S. 457
(1941); Robert Lande & Howard Marvel, The Three Types of Collusion: Fixing Prices,
Rivals, and Rules, 2000 WIS. L. R EV. 941 (2000).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
133
exploitative regulations.
Comparable situations exist elsewhere and antitrust law has
regularly excused such agreements from liability through a
variety of doctrinal labels when the record convincingly showed
that the resulting restraints were either authorized by
government or had a de facto authorization.118 There needs to be
some level of authorization from an appropriate government
entity, the conduct at issue must be within the scope of that
authorization, and there must be an appropriate level of
oversight usually by a public agency, but occasionally some
private organization provides such supervision.119 In energy
markets, such agreements can nevertheless impose unnecessary
restraints that frustrate competition.
Where such agreements result from an open process in
which stakeholders with conflicting interests have an effective
120
voice, there is less basis for concern. These agreements are
likely to be market facilitating and not unduly biased against the
interests of any participant. There should still be disinterested
review of those agreements to protect against the danger of some
subset of stakeholders manipulating the initial decision.121 The
harder cases involve situations in which the relevant
stakeholders share a common economic interest that is
antithetical to market facilitation but are the primary parties to
establish the underlying agreement. Their incentive is to adopt
rules that both serve the legitimate goals of such collaboration
and also suppress disfavored modes.122 Because of the technical
nature of the issues and the incentive to exclude unnecessarily
competition, only an expert agency can possibly balance the
overall efficiency gain against the costs to the market process.
But even if such an agency is strongly pro-competitive, it is
unlikely to give close scrutiny to all the ways in which such
regulations can be deployed to frustrate competition.
In sum, such regulatory agreements, even if subject to strict
oversight, provide both a means directly to undermine the
118.
See Carstensen, supra note 86.
119.
See id.
120.
The standards for such organizations are now codified in the Standard
Development Organization Advancement Act of 2004, H.R. 1086, 108th Cong. (2004)
(amending 15 U.S.C. § 4301 (2005)).
121.
See Allied Tube & Conduit Corp. v. Indian Head, Inc., 486 U.S. 492 (1988); Am.
Soc’y of Mech. Eng’rs v. Hydrolevel Corp., 456 U.S. 556 (1982).
122.
When telephone service was being opened to competition, AT&T attempted to
require all competitors to buy and install costly interface devices between their equipment
and the phone line. Only when the FCC decided that AT&T itself would also have use of
these devices, did AT&T decide that they were not necessary. See Litton Sys., Inc. v.
AT&T, 700 F.2d 785 (2d Cir. 1983).
07CARSTENSEN FORMATTED
134
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
development of a wholesale energy market and a forum for
dominant firms to coordinate competition on matters outside the
scope of such regulation. Here again, the structure of the
markets, especially the electric market with the dominance of
integrated firms, makes it extremely difficult to develop conduct
rules that will both advance the interest in competitive markets
and allow for the necessary level of collective decision making
required to regulate the operation of the system.
E. Ancillary Restraints—Reasonable and Unreasonable
Restraint of trade law also applies to agreements that have
positive, productive objectives. The primary goal of such
agreements is to produce or distribute goods or services.
Inevitably, any forward looking agreement entails some
restrictions on the economic freedom of action of its participants.
Moreover, as parties commit to courses of conduct they become
vulnerable to strategic conduct by other parties to the agreement.
In such contexts, additional limits on future conduct may become
essential to the creation of the agreement or its efficient
operation. The most common justification for such restraints is a
concern for “free-riding” by some parties on sunk investments of
other parties. However, the legitimate concern for strategic
conduct has a more expansive application.
The antitrust concern with restraints found in such
legitimate productive agreements is that they may be unduly
restrictive. Enterprises seek profit and not efficiency. Hence,
there is no inherent constraint on the parties to an agreement
that keeps them from including restraints that facilitate either
unnecessary exploitation of the market or exclude potential
competition beyond that which is immediately relevant to the
venture. In this context, antitrust law has to consider the
legitimacy of the primary objective, the validity of the claimed
justification for any restraint and, assuming some restraint is
justified, whether the specific restraints are no more restrictive
than necessary. There are short cuts in this analysis that arise
from judicial presumptions. Thus, most restraints on distribution
except those setting minimum resale prices are presumed to be
lawful as necessary incidents to the distribution. A court will look
critically at the merits of the restraint only if the party benefiting
from the restraint has substantial market power. In contrast, it
is probably still the case that agreements among competitors that
directly limit their capacity to compete with each other in the
markets other than the one in which they are collaborating are
subject to a presumption of illegality, or at least are subject to
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
135
strict scrutiny.123
The crucial competitive concern is that in the context of
energy markets, long term contracts can have significant
foreclosure effects even though they are also the antidote to short
term market power. The problem is that a generator or a retail
buyer that has significant leverage over the party on the other
side can insist on unduly long or exclusionary terms.
Unintegrated retailers need power, and merchant generators
have substantial sunk costs. Due to of the large extent of vertical
integration, unintegrated firms are marginalized and only
sometimes can deal directly with each other. As a result, when
the integrated electric company agrees to buy power or sell it, it
has a great deal of leverage in the market if the counter party
has few options. Again, this is why integrated firms have little
incentive to remedy the congestion problem.
From a regulatory standpoint, the problem is to define rules
that will limit incentives of dominant buyers (or sellers) to
exploit their power in terms of restraints on the other party’s
freedom to seek or use alternatives. For the reasons already
discussed, the present structure of the market, combined with
the underlying exogenous conditions of power generation, make
this a very difficult process. The structure requires either a very
case specific analysis in which the agency must renegotiate the
agreement to minimize the harms to competition; or a set of firm
rules that may be counterproductive in some number of cases.
But at the same time, especially as these markets return to the
use of longer term contracts, it is essential to the goal of workable
competition that such oversight and control occur to remedy the
abuse of market power that will otherwise transpire.
As with the other areas discussed, the greater potential
problems exist in the electric markets. The gas market has fewer
constraints, more buyers and sellers, and has a history of
contractual supply agreements that provide a base line for
evaluating the merits of any new contract. Such review is not
necessarily helped by FERC’s policy of general acceptance of
contracts.124 Given the need to transform market conduct, FERC
should, to the extent its jurisdiction permits, critically review all
long term supply contracts with special reference to restrictions
on the buyer’s freedom to buy or sell to others, as well as
restrictions on the freedom to resell the energy being
123.
See also Arizona v. Maricopa Medical Found., 457 U.S. 322 (1985); Nat’l Soc’y of
Prof’l Eng’rs v. U.S., 435 U.S. 679, 692 (1978).
124.
See San Diego Gas & Elec. Co. v. FERC, 904 F.2d 727 (D.C. Cir. 1990).
07CARSTENSEN FORMATTED
136
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
purchased.125
F. Jurisdiction and Penalties—Sources and Severity
One of the lessons of antitrust is that multiple enforcers
make it more likely that the commands of the law will be
126
In energy, unlike antitrust, there is a marked
obeyed.
separation between the jurisdiction of the states and the federal
regulatory body. Moreover, under the filed rate doctrine and
exclusive jurisdiction, the regulators have largely, but not
entirely, pre-empted other legal regimes that might impose more
liability on those enterprises that interfere with the development
of the market. The contrast between the world of antitrust and
energy regulation is stark.
Antitrust law enforcement involves multiple actors, in both
public and private sectors. At the federal level, both the Antitrust
Division of the Department of Justice127 and Federal Trade
Commission have authority to enforce the law.128 A key difference
between the two agencies is that only the Antitrust Division has
authority to bring criminal proceedings.129 Criminal cases focus
exclusively on overt cartels and can now result in substantial jail
time for corporate executives as well as substantial fines for
corporate offenders.
Most states have their own antitrust laws with civil and
criminal provisions. In addition, the states have standing to
130
invoke the civil aspect of federal law. This state power provides
an important public interest check on federal agency decisions.
Historically, there have been periods when state litigation has
been very important in maintaining an active role for antitrust.
In addition, private citizens can sue for damages or injunctions
under the antitrust laws.131 There are standing and causation
issues that circumscribe who can sue with respect to any
particular injury, but the threat of treble damages is a widely
known risk that is thought to have significant deterrence value.
The primary function of antitrust law as a market constituting
125.
Ironically, FERC has been trying hard to avoid reviewing the merits of
contracts.
126.
The Supreme Court majority in Trinko, on the other hand, expressed strong
hostility to having multiple enforcers of rules of conduct in the context of regulated
telecommunications. See Trinko, supra note 55, at 413–15.
127.
15 U.S.C. § 4 (2005).
128.
15 U.S.C. § 45 (2005).
129.
15 U.S.C. § 1 (2005).
130.
15 U.S.C. § 15c(a) (2005).
131.
15 U.S.C. § 15(a)1 (2005).
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
137
and regulating force is in its deterrent value.132 The risks of
antitrust liability create incentives for rational managers to
avoid conduct that might produce risks for the enterprise and in
the case of cartel type behavior risks of personal sanctions
including significant jail time. While no system of deterrence is
perfect, it seems plausible that the sanctions that antitrust
imposes have generally lead businesses to avoid overtly
anticompetitive conduct.
There is, of course, an ongoing struggle to define the
contours of permitted and impermissible conduct. A common
complaint is that antitrust law produces “false positives” that
deter efficient conduct.133 A contrary concern is that “false
negatives” authorize anticompetitive conduct that results in
continuing harm to the economy as a result of unnecessary
exclusion or exploitation. Contemporary judicial concerns have
largely focused on the false positives because of a belief that they
pose the greater long term risk.134 Such concerns ignore the
innovative capacity of the market to find alternative routes to
efficient outcomes. The greater concern might be for false
negatives that may well authorize more effective exclusion or
exploitation of markets, thus undermining the long run dynamic
of the market and slowing the restoration of competitive
conditions.
Despite these limitations, antitrust law is an active force
exactly because it can and is being enforced by a variety of
different actors, and if there is a violation, the consequences are
substantial. This is necessary in order to achieve deterrence.
Antitrust enters, except in the case of mergers, only after the
fact. It punishes illegal conduct. This provides the backdrop for
business decisions that must take account of antitrust risks.
The oversight of energy markets remains stuck with the
model of controls that were designed for a regulated market in
which the agency determined price and service. In such a
context, there was less need for after-the-fact sanctions. The goal
was to decide on prices or other elements of the market before
any transactions took place. Both the filed rate doctrine and
exclusive jurisdiction helped to ensure that the regulatory agency
dominated the oversight of the industry. Exclusive jurisdiction
pre-empted state regulation and also foreclosed use of antitrust
law to challenge anti-competitive conduct that the agency had
132.
133.
134.
Tarasi v. Pittsburgh Nat’l Bank, 555 F.2d 1152, 1162 (3rd Cir. 1977).
See Frank Easterbrook, The Limits of Antitrust, 63 TEX . L. REV. 1, 16–17 (1984).
See, e.g., Trinko, supra note 55.
07CARSTENSEN FORMATTED
138
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
approved.135 Despite Otter Tail’s opening of the door to greater
antitrust oversight,136 the FPC, and later FERC, have largely
excluded antitrust as an independent force in the oversight of
these markets.
The regulatory structure for overseeing energy markets is
not adapted to the needs of competition. First, the division
between FERC and the states creates gaps, e.g., integrated
generator-retailers within a state can avoid FERC oversight
despite being major factors in the wholesale side of the energy
market. Second, neither FERC nor the states have the legal tools
to deal effectively with a number of the competitive issues that
exist. FERC lacks the authority over some market participants;
cannot impose severe sanctions for violations of its rules; and
cannot provide the kind of damages that would serve as
appropriate deterrents to misconduct by the firms it insulates
from other legal sanctions.137
The major constraints on the allocation of regulatory
authority make it extremely difficult to fashion workable rules
and create a national market in any type of energy. The limited
jurisdiction of FERC in respect to both gas and electricity
markets is a central problem. FERC lacks the authority to
fashion a workable downstream set of wholesale buyers. Only
state regulation will create that stage of the process. The
artificial wholesale-retail distinction in electricity combined with
the interest that states have in retaining jurisdiction over
important sources of power generation creates serious legal
obstacles to any comprehensive set of conduct-based rules that
would promote competition.
G. Conduct Remedies and the Present State of the Market and
the Law
The foregoing discussion was originally intended to make a
positive contribution to the process of finding appropriate
conduct-based remedies for the major problems of market power
and its abuse that hover over energy markets. As is apparent
from the text that has emerged, there is no basis for optimism
with respect to electric markets given the present structure of
135.
Transmission Agency v. Sierra Pac. Power Co., 295 F.3d 918, 929 (9th Cir.
2002).
136.
Otter Tail Power Co. v. United States, 410 U.S. 366 (1973).
137.
Moreover, even if FERC has jurisdiction, it may lack the express legal authority
to impose a remedy. Basically, it can only condition its approval on a utility agreeing to do
things, but the utility can refuse to accept the condition provided the utility is willing to
forego the benefit of the requested authorization.
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
139
those markets generally and the inherent needs of operating an
integrated electric transmission and distribution system. The
underlying market situations make competition a viable option
only in some situations and only at substantial cost. Whether the
cost is worth the benefit is an open question.
It is, on the other hand, possible to be somewhat more
hopeful about gas markets. The exogenous conditions of that
industry and the structure of the enterprises comprising it make
it possible to retain a more workably competitive wholesale
market. Even there, structural choices have created more risk to
competition than is necessary, leaving little basis to think that
there will be any significant efficiency gains.
V. PREREQUISITES FOR WORKABLE WHOLESALE ENERGY
MARKETS—A RECAPITULATION
The conditions of supply, demand, technology, and
institutional structure all evolved in light of the organization of
the two energy industries as ones in which direct regulation of
prices and services were central organizing principles. As a result
of the combination of exogenous and endogenous factors, both the
structure and conduct of these industries evolved in ways
consistent with those underlying conditions.
Despite some major policy errors, such as trying to impose
conventional rate regulation on wellhead gas prices,138 the trend
of such regulation was to move such markets toward efficiency.
The suggestion is that there were enough pressures on major
producers and retailers of energy from the broader market as
well as regulatory oversight to cause the system to work without
gross inefficiency. The pressures were stronger in gas, where
more market engagement occurred, and weaker in electricity,
where regional disparities were greater because of the tighter
vertical integration and the lack of effective long distance
transmission of lower cost power.
To change to a competitive market system requires careful
consideration of the conditions, structure, and conduct necessary
for desirable performance of a competitive market. To have
workable wholesale markets, it is essential to have sufficient
buyers and sellers interacting in a relatively open context. A
central part of the necessary transformation is disintegration of
the stages of production and distribution. In addition, market
institutions need to be created or expanded, and the kinds of
transactions on which the market will rely must be identified and
138.
See supra text accompanying note 8.
07CARSTENSEN FORMATTED
140
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
legally defined. This is a market creation and facilitation process
that is akin to moving an economy from state socialism to a
market basis.139 It cannot be accomplished by slogans or
simplistic legislation.
The contrast between the gas and electric markets is
striking. The basis for the differences reside in both the
exogenous conditions of the two markets and the structures as
well as conduct that evolved based on those conditions, and the
endogenous historical facts and legal policies applied to the two
industries. The institutional, regulatory, supply, and demand
characteristics of natural gas made a transition to more
competition easier to visualize even if it posed serious political
challenges. A market already operated at the production end of
the natural gas chain. The gas pipelines were not substantially
integrated into retail distribution, and a number of major buyers
existed in both the unregulated intrastate market as well as had
the capacity to enter into the interstate market when permitted
to do so.
Thus, conditions, structure, and conduct all made a
transition to a broader competitive gas market feasible. The
central changes in the basic conditions of the industry were to
separate transportation charges and responsibility from the
buying and selling of gas itself. This included the creative idea of
converting rights to use pipeline capacity into tradable units that
could be bought and sold. As discussed earlier, this conversion
limited the incentives of the pipeline owner to manipulate
capacity and seek to impose monopoly prices. Of course,
regulation of price remains essential for the delivery services
because those remain potential monopolies, especially at the ends
of many pipelines where the ultimate buyers have only one way
to receive gas.
The most questionable element of this market is the decision
to allow pipeline owners to remain merchants of gas sold through
their affiliated pipelines. Conduct-oriented rules imposed on this
activity attempt to reduce or eliminate the manifest incentive to
manipulate supply in order to raise prices. As discussed earlier, a
structural response, similar to that adopted in other
transportation service industries that would ban such activity
because of the inherent conflict would seem the better approach.
Absent such a separation, regulators need to have the power to
139.
An interesting comparative institutional point is that in countries where the
government owned most or all generation, transmission, and distribution, the transition
to market systems was easier because the state could more freely decide how to configure
ownership structures as it sold off its assets.
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
141
ban any merchant from the market if that merchant engages in
manipulative or deceptive acts or practices. These requirements
are common to the market facilitating regulation of other major
commodities markets. Deterrence is important with respect to all
traders, but especially important when traders have an inherent
conflict of interest.
The other continuing structural concern for wholesale
markets in gas is the need to have policies that will retain
sufficient numbers of buyers and sellers to ensure that overall
market participants have a strong and clear incentive to seek fair
and open market practices. Thus, stricter standards for merger
on both the buying and selling side of the market seem in order.
In addition, there should be concern with any combination that
results in incentives to manipulate capacity of pipelines. In
particular, the concern is with combinations of pipeline
ownership and ownership of electric generation facilities in the
area served by such a pipeline.
In the case of electricity, the problems of moving to a more
competitive market are far more substantial. Indeed, absent
significant structural change and changes in the endogenous
conditions of electric markets, the goal of workably competitive
wholesale markets in electricity seems problematic at best.140 The
essential structural condition for competition is a separation of
the ownership of generation from that of transmission and
dispersing both generation and distribution among sufficient
owners to create a context in which a wholesale market can
flourish. Unlike gas, the technological conditions for generation,
transmission, and distribution require much closer and
continuous interaction among the parties. This poses major
institutional design problems to ensure responsibility for
balancing and reliability as well as reorientation of the delivery
system, transmission, and distribution, to serve a competitive
supply market. The examples of securities and commodities
markets tells us that such institutional designs are feasible, but
in the case of electricity, they would have to be built from the
ground up. In addition, the actual physical layout of the
transmission system was not intended to serve as a medium for
market exchanges. Thus, a major investment in new
transmission capacity with all the costs and environmental
concerns entailed is necessary. Finally, the regulatory conditions
140.
See Andrea Morris, Why the Music is Off-Key When Lawyers Sing from
Economists’ Songbooks or Why Public Utility Deregulation Will Fail, in THE END OF
NATURAL MONOPOLY: DEREGULATION AND COMPETITION IN THE ELECTRIC POWER
INDUSTRY, 193 (Peter Grossman & Daniel Cole, ed. 2003).
07CARSTENSEN FORMATTED
142
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
under which electric markets operate need to be transformed to
ones appropriate to a competitive market system. The current
divisions of responsibility, the exclusivity of the several
regulators, and the limits on the sanctions allowed to them are
not the necessary conditions for the creation of a workably
competitive wholesale market in electricity.
Specific questions of market design enter into the
evaluation. Because of the major linkages between generation
and distribution, there is a heavy emphasis on short-term
transactional markets to cover the marginal supplies. In
California, all supplies were bought on a short-term basis
creating a mismatch between incentives for development of new
and more efficient generation and the market for such power
because of the long-term risks in construction having only a short
term market.
The problem of market design, i.e., conduct, is a very
difficult one as FERC has discovered. Because of the physics of
electricity, it is difficult for parties to transact in actual electrons.
What is required is a method of matching purchases and
withdrawals from the system. Comparable settlement systems
exist in commodities and securities trading markets. Those
markets operate with a strong central authority in the position of
market maker. That entity must police the conduct of buyers and
sellers to ensure that all trades are completed. The operator of
the transmission system stands in a comparable position in
electricity. That operator provides the central clearing point for
supply and demand. If the overall transmission system has
sufficient capacity, including the ability to use power at a variety
of points, it can match contractual rights to deliver power with
the demand for power. This is similar to the clearing function of
commodities and securities markets. What is then required is a
set of contractual instruments that provide long-term as well as
short-term commitments from generators that distributors can
buy and deliver to the system operator to cover their
requirements. Because the value of electricity varies by time of
day and season of the year, it is much harder to develop the
instruments that can be bought and sold in sufficient volume to
create a market.141 It would appear that longer term contracts
that involve continued commitments to supply some set level of
power, equivalent to the “take or pay” contracts used in gas
142
markets at one time, may be potentially useful tools.
141.
The industry has been trying to develop various financial instruments to
accomplish this purpose.
142.
See BOSSELMAN, supra note 7, at 468–73.
07CARSTENSEN FORMATTED
3/31/2006 9:24:04 AM
2006] CREATING WORKABLY COMPETITIVE MARKETS
143
It should have been clear to policy makers at the outset and
has certainly become clear in hindsight that positive results from
movement toward competition in electric markets will not come
easily because of the complexity of the transformations required.
Yet so deep was the political faith in “competition” that
legislation such as California’s imposed unworkable legal
structures on the industry.143 These policy errors in turn call into
question the claims that competitive markets will provide long
run economic benefits. Indeed, given the complexity of the
electricity business, either competitive or regulated markets will
operate imperfectly. The central question is whether more
competition will result in a less imperfect market than more
regulation.
To make progress toward a workable competitive market
requires a major reorientation of the physical structure of the
transmission system, its separation from the strategic interests
of its former owners, and the development of market mechanisms
that would allow generators to sell directly to buyers. These
transformations cannot come without major revisions of the legal
conditions under which power is produced, transmitted, and
distributed.144 Despite repeated calls for such legislation,145 it has
not emerged, and the proposals that congress has considered
have fallen substantially short of providing the overall
restructuring necessary to bring about workable markets.
Overall competitive markets are preferable to regulated
ones. The competitive market brings a greater opportunity for
entry, exit, and the potential for more dynamics in technology.
But, such markets are not free and require a great deal of careful
design and continuing re-evaluation. In the case of electricity, the
costs of transition as well as those associated with the ongoing
efforts to oversee a market including the major economic costs
associated with exploitation of market power may well outweigh
any long term benefits that might arise if competition caused
changes in the basic conditions and structure of the market.
Without significant change in conditions and structure, indeed, it
is possible that the present system is the worst of all possible
options.
143.
See Bush & Mayne, supra note 93.
144.
It is not unduly pessimistic to suggest that the current muddle in electricity is
worse than either a fully competitive or a fully regulated system. Much unregulated
market power exists in the current system that can be and is exploited under the guise of
competition. Yet the bifurcated regulatory framework of federal and state jurisdiction
with its highly artificial allocation of authority and the lack of effective sanctions against
many of the market abuses that have occurred contribute to the muddle.
145.
See, e.g., Kelliher, supra note 81.
07CARSTENSEN FORMATTED
144
3/31/2006 9:24:04 AM
ENVIRONMENTAL & ENERGY LAW & POLICY J.
[1:1
To paraphrase Orwell, “all markets are equal, but some
markets are more equal than others.” The central lessons from a
comparison of the effort to create workable wholesale markets in
gas and electricity is that conditions and existing structures are
very important factors. Without strong legislative will to make
significant changes that are very market specific, the prospects
for remedying the flaws in poorly functioning markets are
limited. Competition may be desirable, but only when it is
feasible.